7.1.1 Interactive compression for omnidirectional images and texture maps of 3D models

• Keywords:
  Image compression, Random access
• Functional Description:

  This code implements a new image compression algorithm that allows to navigate within a static scene. To do so, the code provides access in the compressed domain to any block and therefore allows extraction of any subpart of the image. This codec implements this interactive compression for two image modalities: omnidirectional images and texture maps of 3D models. For omnidirectional images the input is a 2D equirectangular projection of the 360 image. The output is the image seen in the viewport. For 3D models, the input is a texture map and the 3D mesh. The output is also the image seen in the viewport.

  The code consists of three parts: (A) an offline encoder (B) an online bit extractor and (C) a decoder. The offline encoder (i) partitions the image into blocks, (ii) optimizes the positions of the access blocks, (iii) computes a set of geometry aware predictions for each block (to cover all possible navigation paths), (iv) implements transform quantization for all blocks and their predictions, and finally (v) evaluates the encoding rates. The online bit extractor (Part B) first computes the optimal and geometry aware scanning order. Then it extracts in the bitstream, the sufficient amount of information to allow the decoding of the requested blocks. The last part of the code is the decoder (Part C). The decoder reconstructs the same scanning order as the one computed at the online bit extractor. Then, the blocks are decoded (inverse transform, geometry aware predictions, ...) and reconstructed. Finally the image in the viewport is generated.

• Authors:
  Navid Mahmoudian Bidgoli, Thomas Maugey, Aline Roumy
• Contact:
  Aline Roumy

7.1.2 GRU_DDLF

• Name:
  Gated Recurrent Unit and Deep Decoder based generative model for Light Field compression
• Keywords:
  Light fields, Neural networks, Deep learning, Compression
• Functional Description:
  This code implements a neural network-based light field representation and compression scheme The code is developped in python, based on the pytorch deep learning framework..The proposed network is based on both a generative model that aims at modeling the spatial information that is static, i.e., found in all light field views, and on a convolutional Gated Recurrent Unit (ConvGRU) that is used to model variations between blocks of angular views. The network is untrained in the sense that it is learned only on the light field to be processed, without any additional training data. The network weights can be considered as a representation of the input light field. The compactness of this representation depends on the number of weights or network parameters, but not only. It also depends on the number of bits needed to accurately quantize each weight. Our network is thus learned using a strategy that takes into account weight quantization, in order to minimize the effect of weight quantization noise on the light field reconstruction quality. The weights of the different filters are optimized end-to-end in order to optimize the image reconstruction quality for a given number of quantization bits per weight.
• Contact:
  Christine Guillemot
• Participants:
  Xiaoran Jiang, Jinglei Shi, Christine Guillemot

7.1.3 FPFR+

• Name:
  Fused Pixel and Feature based Reconstruction for view synthesis and temporal interpolation
• Keywords:
  Light fields, View synthesis, Temporal interpolation, Neural networks
• Functional Description:
  This code implements a deep residual architecture that can be used both for synthesizing high quality angular views in light fields and temporal frames in classical videos. The code is developped in python, based on the tensorflow deep learning framework. The proposed framework consists of an optical flow estimator optimized for view synthesis, a trainable feature extractor and a residual convolutional network for pixel and feature-based view reconstruction. Among these modules, the fine-tuning of the optical flow estimator specifically for the view synthesis task yields scene depth or motion information that is well optimized for the targeted problem. In cooperation with the end-to-end trainable encoder, the synthesis block employs both pixel-based and feature-based synthesis with residual connection blocks, and the two synthesized views are fused with the help of a learned soft mask to obtain the final reconstructed view.
• Contact:
  Christine Guillemot
• Participants:
  Jinglei Shi, Xiaoran Jiang, Christine Guillemot

7.1.4 PnP-A

• Name:
  Plug-and-play algorithms
• Keywords:
  Algorithm, Inverse problem, Deep learning, Optimization
• Functional Description:
  The software is a framework for solving inverse problems using so-called "plug-and-play" algorithms in which the regularisation is performed with an external method such as a denoising neural network. The framework also includes the possibility to apply preconditioning to the algorithms for better performances. The code is developped in python, based on the pytorch deep learning framework, and is designed in a modular way in order to combine each inverse problem (e.g. image completion, interpolation, demosaicing, denoising, ...) with different algorithms (e.g. ADMM, HQS, gradient descent), diifferent preconditioning methods, and different denoising neural networks.
• Contact:
  Christine Guillemot
• Participants:
  Mikael Le Pendu, Christine Guillemot

7.1.5 SIUPPA

• Name:
  Stochastic Implicit Unrolled Proximal Point Algorithm
• Keywords:
  Inverse problem, Optimization, Deep learning
• Functional Description:
  This code implements a stochastic implicit unrolled proximal point method, where an optimization problem is defined for each iteration of the unrolled ADMM scheme, with a learned regularizer. The code is developped in python, based on the pytorch deep learning framework. The unrolled proximal gradient method couples an implicit model and a stochastic learning strategy. For each backpropagation step, the weights are updated from the last iteration as in the Jacobian-Free Backpropagation Implicit Networks, but also from a randomly selected set of unrolled iterations. The code includes several applications, namely denoising, super-resolution, deblurring and demosaicing.
• Contact:
  Christine Guillemot
• Participants:
  Brandon Le Bon, Mikael Le Pendu, Christine Guillemot

7.1.6 DeepLFCam

• Name:
  Deep Light Field Acquisition Using Learned Coded Mask Distributions for Color Filter Array Sensors
• Keywords:
  Light fields, Deep learning, Compressive sensing
• Functional Description:
  This code implements a deep light field acquisition method using learned coded mask distributions for color filter array sensors. The code describes and allows the execution of an algorithm for dense light field reconstruction using a small number of simulated monochromatic projections of the light field. Those simulated projections consist of the composition of: - the filtering of a light field by a color coded mask placed between the aperture plane and the sensor plane, performing both angular and spectral multiplexing, - a color filtering array performing color multiplexing, - a monochromatic sensor. This composition of filtering, sampling, projection, is modeled as linear projection operator. Those measurements, along with the 'modulation field', or 'shield field' corresponding to the acquisition device (i.e. the light field corresponding to the the transformation of a uniformly white light field, known to completelly characterize the projection operator), are then feed to a deep convolutional residual network to reconstruct the original light field. The provided implementation is meant to be used with Tensorflow 2.1.
• Publication:
  leguludec:hal-03203347
• Contact:
  Guillaume Le Guludec
• Participants:
  Guillaume Le Guludec, Christine Guillemot

7.1.7 DeepULFCam

• Name:
  Deep Unrolling for Light Field Compressed Acquisition using Coded Masks
• Keywords:
  Light fields, Optimization, Deep learning, Compressive sensing
• Functional Description:
  This code describes and allows the execution of an algorithm for dense light field reconstruction using a small number of simulated monochromatic projections of the light field. Those simulated projections consist of the composition of: - the filtering of a light field by a color coded mask placed between the aperture plane and the sensor plane, performing both angular and spectral multiplexing, - a color filtering array performing color multiplexing, - a monochromatic sensor. The composition of these filterings/projects is modeled as linear projection operator. The light field is then reconstructed by performing the 'unrolling' of an interative reconstruction algorithm (namely the HQS 'half-quadratic splitting' algorithm, a variant of ADMM, an optimization algorithm, applied to the solving a regularized least-squares problem) into a deep convolutional neural network. The algorithm makes use of the structure of the projection operator to efficiently solve the quadratic data-fidelty minimization sub-problem in closed form. This code is designed to be compatible with Python 3.7 and Tensorflow 2.3. Other required dependencies are: Pillow, PyYAML.
• Contact:
  Guillaume Le Guludec
• Participants:
  Guillaume Le Guludec, Christine Guillemot

7.1.8 OSLO: On-the-Sphere Learning for Omnidirectional images

• Keywords:
  Omnidirectional image, Machine learning, Neural networks
• Functional Description:
  This code implements a deep convolutional neural network for omnidirectional images. The approach operates directly on the sphere, without the need to project the data on a 2D image. More specifically, from the sphere pixelization, called healpix, the code implements a convolution operation on the sphere that keeps the orientation (north/south/east/west). This convolution has the same complexity as a classical 2D planar convolution. Moreover, the code implements stride, iterative aggregation, and pixel shuffling in the spherical domain. Finally, image compression is implemented as an application of this on-the-sphere CNN.
• Authors:
  Navid Mahmoudian Bidgoli, Thomas Maugey, Aline Roumy
• Contact:
  Aline Roumy

7.1.9 JNR-MLF

• Name:
  Joint Neural Representation for Multiple Light Fields
• Keywords:
  Neural networks, Light fields
• Functional Description:

  The code is designed to learn a joint implicit neural representation of a collection of light fields. Implicit neural representations are neural networks that represent a signal as a mapping from the space of coordinates to the space of values. For traditional images, it is a mapping from a 2D space to a 3D space. In the case of light field, it is a mapping from a 4D to a 3D space.

  The algorithm works by learning a factorisation of the weights and biases of a SIGNET based network. However, unlike SIGNET which uses the Legendre polynomials as the positional encoding, the model used in the code is using learned Fourier Features.

  For each layer, a base of matrices is learned that serves as a representation shared between all light fields (a.k.a. scene) of the dataset, together with, for each scene in the dataset, a set of coefficients with respect to this base, which acts as individual representations. More precisely, the code uses matrix factorization inspired from the Singluar Value Decomposition (SVD).

  The matrices formed by taking the linear combinations of the base matrices with the coefficients corresponding to a given scene serve as the weight matrices of the network. In addition to the set of coefficients, we also learn an individual bias vector for each scene.

  The code is therefore composed of two parts: - The representation provider, which takes the index i of a scene, and outputs a weight matrix and bias vector for each layer. - The synthesis network which, using the weights and biases, computes the values of the pixels by querying the network on the coordinates of all pixels.

  The network is learned using Adam, a flavour of the SGD algorithm. At each iteration, a light field index i is chosen at random, along with a batch of coordinates, and the corresponding values are predicted, using weights and biases from the representation provider and the synthesis network. The model's parameters U, V and (the relevant column of) Sigma are then updated by gradient backprogagation from a MSE minimization objective. The code uses TensorFlow 2.7.

• Contact:
  Guillaume Le Guludec
• Participant:
  Guillaume Le Guludec

7.2.1 CLIM processing toolbox

Participants: Christine Guillemot, Laurent Guillo.

As part of the ERC Clim project, the EPI Sirocco is developing a light field processing toolbox. The toolbox and libraries are developed in C++ and the graphical user interface relies on Qt. As input data, this tool accepts both sparse light fields acquired with High Density Camera Arrays (HDCA) and denser light fields captured with plenoptic cameras using microlens arrays (MLA). At the time of writing, in addition to some simple functionalities, such as re-focusing, change of viewpoints, with different forms of visualization, the toolbox integrates more advanced tools for scene depth estimation from sparse and dense light fields, for super-ray segmentation and scene flow estimation, and for light field denoising and angular interpolation using anisotropic diffusion in the 4D ray space. The toolbox is now being interfaced with the C/C++ API of the tensorflow platform, in order to execute deep models developed in the team for scene depth and scene flow estimation, view synthesis, and axial super-resolution.

7.2.2 ADT: Interactive Coder of Omnidirectional Videos

Participants: Sébastien Bellenous, Navid Mahmoudian Bidgoli, Thomas Maugey, Aline Roumy.

By nature, a video 360∘^\circ cannot be seen entirely, because it covers all the direction. The developed coding scheme will aim at 1) transmitting only what is necessary and 2) bringing no transmission rate overhead

In the Intercom project, we have studied the impact of interactivity on the coding performance. We have, for example, tackled the following problem: is it possible to compress a 360 video (as shown in Fig.1) once for all, and then partly extract and decode what is needed for a user navigation, while, keeping good compression performance? After having derived the achievable theoretical bounds, we have built a new omnidirectional video coder. This original architecture enables to reduce significantly the cost of interactivity compared to the conventional video coders. The project ICOV proposes to start from this promising proof of concept and to develop a complete and well specified coder that is aimed to be shared with the community.

8.1.1 Light Field Acquisition Using Colour Coded Masks and Meta-Surfaces

Participants: Anil Ipek Atalay Appak, Christine Guillemot, Guillaume Le Guludec.

Compressed sensing using color-coded masks has been recently considered for capturing light fields using a small number of measurements. Such an acquisition scheme is very practical, since any consumer-level camera can be turned into a light field acquisition camera by simply adding a coded mask in front of the sensor. We developed an efficient and mathematically grounded deep learning model to reconstruct a light field from a set of measurements obtained using a color-coded mask and a color filter array (CFA). Following the promising trend of unrolling optimization algorithms with learned priors, we formulate our task of light field reconstruction as an inverse problem and derive a principled deep network architecture from this formulation 11. We also introduce a closed-form extraction of information from the acquisition, while similar methods found in the recent literature systematically use an approximation. Compared to similar deep learning methods, we show that our approach allows for a better reconstruction quality. We further show that our approach is robust to noise using realistic simulations of the sensing acquisition process.

We pursue this study by focusing on microscopy with extended depth of field. Indeed, in conventional microscopy imaging, it is not possible to clearly see all objects at different depths due to optics limitations. Extended Depth of Field (EDOF) lenses can solve this problem, however there is a trade-off between the resolution and the depth of field. To overcome this trade-off, phase masks have been used combined them with reconstruction networks.

In collaboration with the university of Tampere (Prof. Humeyra Caglayan, Erdem Sahin) in the context of the Plenoptima Marie Curie project, we adress this problem of EDOF while preserving a high resolution by designing a pipeline for end-to-end learning and optimization of an optical layer together with deep learning based image reconstruction methods. In our design, instead of phase masks, we consider instead meta-surface materials. Meta-surfaces are periodic sub-wavelength structures that can change the amplitude, phase, and polarization of the incident wave. Its parameters are end-to-end optimized to have the highest depth of field with good spatial resolution and high reconstruction quality.

While, in a first step we focused on 2D microscopy imaging, the next step will be to consider light field microscopy. This co-design should indeed lead to a new plenoptic camera design which would take full advantage of the sensor resolution and allow for extended depth of fields, a property that is very important in microscopy.

8.1.2 Sparse Non Parametric BRDF Model and Plenopic Point Clouds

Participants: Davi Freitas, Christine Guillemot.

In collaboration with the Univ. of Linkoping, we have derived a novel sparse non-parametric BRDF model using a machine learning approach to represent the space of possible BRDFs using a set of multidimensional sub-spaces, or dictionaries 16. By training the dictionaries under a sparsity constraint, the model guarantees high quality representations with minimal storage requirements and an inherent clus- tering of the BDRF-space. The model can be trained once and then reused to represent a wide variety of measured BRDFs. Moreover, the proposed method is flexible to incorporate new unobserved data sets, parameterizations, and transformations. In addition, we show that any two, or more, BRDFs can be smoothlyinterpolated in the coefficient space of the model rather than the significantly higher-dimensional BRDF space. Experimental results show that the proposed approach results in about 9.75dB higher SNR on average for rendered images as compared to current state-of-the-art models.

We pursue this study in the context of the Plenoptima project by designing a novel scene representation, a richer model, in which color would be untangled from lighting, using BRDF models. In explicit representations like plenoptic point clouds (PPC), the RGB colors in the model indeed result from interactions between the incoming light during capture and the material and shape properties of the object in the scene. Hence, if the scene modelled by the PPC is put into an environment with different lighting conditions, it would look out of place since these conditions at the time of capture are “baked in” the colors of the PPC. To address this problem, we leveraged the plenoxel representation and developed a plenoxel point cloud representation in which view-dependent spatial information, represented by BRDF and compressed with spherical harmonics, is associated to the voxel grid.

8.1.3 A Light Field FDL-HCGH Feature in Scale-Disparity Space

Participants: Christine Guillemot.

Many computer vision applications rely on feature detection and description, hence the need for computationally efficient and robust 4D light field (LF) feature detectors and descriptors. In collaboration with Xi'an University (Prof. Zhaolin Xiao), we have proposed a novel light field feature descriptor based on the Fourier disparity layer representation, for light field imaging applications 18. After the Harris feature detection in a scale-disparity space, the proposed feature descriptor is then extracted using a circular neighborhood rather than a square neighborhood. It is shown to yield more accurate feature matching, compared with the LiFF LF feature, with a lower computational complexity. In order to evaluate the feature matching performance with the proposed descriptor, we generated a synthetic stereo LF dataset with ground truth matching points. Experimental results with synthetic and real-world datasets show that our solution outperforms existing methods in terms of both feature detection robustness and feature matching accuracy. It is in particular shown to yield more accurate feature matching, compared with the reference LIght Field Feature (LiFF) descriptor, with a lower computational complexity. Experimental results with synthetic and real-world datasets show that our solution outperforms existing methods in terms of both feature detection robustness and feature matching accuracy.

8.2.1 Neural Network Priors for Light Field Representation

Participants: Christine Guillemot, Xiaoran Jiang, Jinglei Shi.

Deep generative models have proven to be effective priors for solving a variety of image processing problems. However, the learning of realistic image priors, based on a large number of parameters, requires a large amount of training data. It has been shown recently, with the so-called deep image prior (DIP), that randomly initialized neural networks can act as good image priors without learning.

We have proposed a deep generative model for light fields, which is compact and which does not require any training data other than the light field itself. The proposed network is based on both a generative model that aims at modeling the spatial information that is static, i.e., found in all light field views, and on a convolutional Gated Recurrent Unit (ConvGRU) that is used to model variations between angular views 10. The spatial view generative model is inspired from the deep decoder, itself built upon the deep image prior, but that we enhance with spatial and channel attention modules, and with quantization-aware learning. The attention modules modulate the feature maps at the output of the different layers of the generator. In addition, we offer an option which expressively encodes the upscaling operations in learned weights in order to better fit the light field to process. The deep decoder is also adapted in order to model several light field views, with layers (i.e. features) that are common to all views and others that are specific to each view. The weights of both the convGRU and the deep decoder are learned end-to-end in order to minimize the reconstruction error of the target light field.

To show the potential of the proposed generative model, we have developed a complete light field compression scheme with quantization-aware learning and entropy coding of the quantized weights. Experimental results show that the proposed method outperforms state-of-the-art light field compression methods as well as recent deep video compression methods in terms of both PSNR and MS-SSIM metrics.

8.2.2 Neural Radiance Fields and Implicit Scene Representations

Participants: Christine Guillemot, Guillaume Le Guludec, Jinglei Shi.

The concept of Neural Radiance Fields (NeRF) has been recently introduced for scene representation and light field view synthesis. NeRF is an implicit Multi-Layer Perceptron-based model that maps 5D vectors—3D coordinates plus 2D viewing directions—to opacity and color values, computed by fitting the model to a set of training views. The resulting neural model can then be used to generate any view of the light field, in a continuous manner, using volume rendering techniques.

NeRF can thus be seen as a neural representation of a scene which is learned from a set of input views. While existing light field compression solutions usually consider encoding and transmitting a subset of light field views which are then used at the decoder to synthesize and render the entire light field, we consider an alternative approach which optimizes and compresses the NeRF on the sender side, so that the network itself is transmitted to the receiver.

For high efficiency, we have developed a Distilled Low Rank optimization method for Neural Radiance Fields. The resulting compact neural radiance fields can then be efficiently used for compressing the light field of a scenes. While existing compression methods encode the set of light field sub-aperture images, our proposed method instead learns an implicit scene representation in the form of a Neural Radiance Field (NeRF), which also enables view synthesis. For reducing its size, the model is first learned under a Low Rank (LR) constraint using a Tensor Train (TT) decomposition in an Alternating Direction Method of Multipliers (ADMM) optimization framework. To further reduce the model size, the components of the tensor train decomposition need to be quantized. However, performing the optimization of the NeRF model by simultaneously taking the low rank constraint and the rate-constrained weight quantization into consideration is challenging. To deal with this difficulty, we introduce a network distillation operation that separates the low rank approximation and the weight quantization in the network training. The information from the initial LR constrained NeRF (LR-NeRF) is distilled to a model of a much smaller dimension (DLR-NeRF) based on the TT decomposition of the LR-NeRF. An optimized codebook is then learned to quantize all TT components, producing the final quantized representation. Experimental results show that our proposed method yields better compression efficiency compared with state-of-the-art methods, and it additionally has the advantage of allowing the synthesis of any light field view with a high quality.

Averaged reconstruction error maps of decompressed light fields, when using the Jpeg-Pleno standard (JPL), the HEVC compression standard, and deep learning-based video compression solutions (RLVC, HLVC). We can see that our low-rank constrained neural radiance field representation gives a better reconstruction quality (error map with lower values, higher PSNR)

Neural implicit representations have thus appeared as promising techniques for representing a variety of signals, among which light fields, offering several advantages over traditional grid-based representations, such as independence to the signal resolution. While, as described above, we have first focused on finding a good representation for a given type of signal, i.e. for a particular light field, we realized that exploiting the features shared between different scenes remains an understudied problem. We have therefore made one step towards this end by developing a method for sharing the representation between thousands of light fields, splitting the representation between a part that is shared between all light fields and a part which varies individually from one light field to another. We show that this joint representation possesses good interpolation properties, and allows for a more light-weight storage of a whole database of light fields, exhibiting a ten-fold reduction in the size of the representation when compared to using a separate representation for each light field.

8.2.3 Omni-directional Neural Radiance Fields

Participants: Kai Gu, Christine Guillemot, Thomas Maugey.

In this study, we explore the use of fish-eye cameras to acquire spherical light fields, the goal being to design novel representations and reconstruction methods from omni-directional imagery. Our motivation is to be able to capture or reconstruct light fields with a very large field of view, i.e. 360°, using fish-eye lenses and developing the corresponding advanced reconstruction and synthesis algorithms. We focus on the question: how do we extract omni-directional information and potentially benefit from it when reconstructing a spherical light field of a large-scale scene with a non- converged camera setup?.

To address this question, we have developed an omni-directional Neural radiance Field estimation method, called omni-NERF 25. While NeRF was originally designed for ideal pin-hole camera models with known camera poses, in omni-NERF a new projection model for the distortion of fisheye lenses as a function between the incident and refracted rays has been defined. The model can reconstruct a 360 light field and can fit different fisheye projection models and optimize the camera parameters jointly with the radiance field. We have shown that the spherical sampling improves the performance when training with panoramic fisheye or wide-angle perspective images, and that it improves the performance of view synthesis compared to planar sampling of the scene.

8.2.4 Satellite image compression and restoration

Participants: Denis Bacchus, Christine Guillemot, Arthur Lecert, Aline Roumy.

In the context of the Lichie project, in collaboration with Airbus, we address two problems for satellite imaging: quasi-lossless compression and restoration, using deep learning methods.

Designing efficient compression algorithms for satellite images must take into account several constraints. First, (i) it must be well adapted to the raw data format. In particular, we consider in this work, the cameras for the Lion satellite constellation with ultra-high spatial resolution at the price of a lower spectral resolution. More precisely, the three spectral bands (RGB) are acquired with a single sensor with an in-built filter array. Second, (ii) it should be adapted to the image statistics. Indeed, satellite mages contain very high-frequency details with small objects spread over very few pixels only. Finally, (iii) the compression must be quasi-lossless to allow accurate on-ground interpretation. We have proposed a compression algorithm, based on an autoencoder, that can efficiently deal with these three constraints.

Indeed, while existing compression solutions mostly assume that the captured raw data has been demosaicked prior to compression, we have designed an end-to-end trainable neural network for joint compression and demosaicking of satellite images. We have first introduced a training loss combining a perceptual loss with the classical mean square error, which has been shown to better preserve the high-frequency details present in satellite images. We then developed a multi-loss balancing strategy which significantly improves the performance of the proposed joint demosaicking-compression solution.

To restore the low light images, we have designed a GAN-based solution to decompose the low light images into its components according to the retinex model (illumination, reflectance). We have introduced different regularization constraints when learning the Retinex model components and we have explored different variants of the retinex model, either assuming a monochomatic or a coloured illumination component. The approach yields state of the art solutions while being very fast for the restoration.

8.2.5 Neural networks for video compression acceleration

Participants: Christine Guillemot, Yiqun Liu, Aline Roumy.

In the context of the Cifre contract with Ateme we investigate deep learning architectures for the inference of coding modes in video compression algorithms with the ultimate goal of reducing the encoder complexity. We focused on the acceleration of the Versatile Video Coding (VVC) standard.

The Versatile Video Coding (VVC) standard has been finalized in 2020. Compared to the High Efficiency Video Coding (HEVC) standard, VVC offers about 50% compression efficiency gain, at the cost of about 10x more encoder complexity. We have developed a Convolutional Neural Network (CNN)-based method to speed up the inter partitioning in VVC 27. Our method operates at the Coding Tree Unit (CTU) level, by splitting each CTU into a fixed grid of 8×8 blocks. Then each cell in this grid is associated with information about the partitioning depth within that area. A lightweight network for predicting this grid is employed during the rate-distortion optimization to avoid partitions that are unlikely to be selected. Experiments show that the proposed method can achieve acceleration ranging from 17% to 30% in the Random Access Group Of Picture 32 (RAGOP32) mode of VVC Test Model (VTM)10 with a reasonable efficiency drop ranging from 0.37% to 1.18% in terms of BD-rate increase.

8.3.1 Plug-and-play optimization and learned priors

Participants: Rita Fermanian, Christine Guillemot, Mikael Le Pendu.

Recent optimization methods for inverse problems have been introduced with the goal of combining the advantages of well understood iterative optimization techniques with those of learnable complex image priors. A first category of methods, referred to as ”Plug-and-play” methods, has been introduced where a learned network-based prior is plugged in an iterative optimization algorithm. These learnable priors can take several forms, the most common ones being: a projection operator on a learned image subspace, a proximal operator of a regularizer or a denoiser.

In the context of the AI chair DeepCIM, we have first studied Plug-and-Play optimization methods for solving inverse problems by plugging a denoiser into the ADMM (Alternating Direction Multiplier Method) optimization algorithm. We have extended the concept of plug-and-play optimization to use denoisers that can be parameterized for non-constant noise variance. In that aim, we have introduced a preconditioning of the ADMM algorithm, which mathematically justifies the use of such an adjustable denoiser 12. We additionally proposed a procedure for training a convolutional neural network for high quality non-blind image denoising that also allows for pixel-wise control of the noise standard deviation. We have shown that our pixel-wise adjustable denoiser, along with a suitable preconditioning strategy, can further improve the plug-and-play ADMM approach for several applications, including image completion, interpolation, demosaicing and Poisson denoising. An illustration of Poisson denoising results is given in Fig.(4).

Poisson Denoising for very high noise level. N is the number of iterations of the ADMM iterative algorithms

We have further shown that it is possible to train a network directly modeling the gradient of a MAP regularizer while jointly training the corresponding MAP denoiser 8. We use this network in gradient-based optimization methods and obtain better results comparing to other generic Plug-and-Play approaches. We also show that the regularizer can be used as a pre-trained network for unrolled gradient descent. Lastly, we show that the resulting denoiser allows for a better convergence of the Plug-and-Play ADMM.

8.3.2 Unrolled Optimization and deep equilibrium models

Participants: Christine Guillemot, Brandon Le Bon, Mikael Le Pendu, Samuel Willingham.

One advantage of plug-and-play methods with learned priors is their genericity in the sense that they can be used for any inverse problem, and do not need to be re-trained for each new problem. However, priors learned independently of the targeted problem may not yield the best solution. Unrolling a fixed number of iterations of optimization algorithms is another way of coupling optimization and deep learning techniques. The learnable network is trained end to- end within the iterative algorithm so that performing a fixed number of iterations yields optimized results for a given inverse problem.

While iterative methods usually iterate until idempotence, the number of iterations in unrolled optimization methods is set to a small value. This makes it possible to learn a component end-to-end within the optimization algorithm, hence in a way which takes into account the data term, i.e., the degradation operator. But learning networks end-to-end within an unrolled optimization scheme requires high GPU memory usage since the memory used for the backpropagation scales linearly with the number of iterations. This explains why the number of iterations used in an unrolled optimization method is limited. To cope with these limitations, we have developed a stochastic implicit unrolled proximal point algorithm with a learned denoiser, in which sub-problems are defined per iteration. We exploit the fact that the Douglas-Rachford algorithm is an application of the proximal point algorithm to re-define the unrolled step as a proximal mapping. We focused on the unrolled ADMM, which has been demonstrated to be a special case of the Douglas-Rachford algorithm, hence of the proximal point algorithm. This allows us to introduce a novel unrolled proximal gradient method coupling an implicit model and a stochastic learning strategy. We have shown that this stochastic iteration update strategy better controls the learning at each unrolled optimization step, hence leads to a faster convergence than other implicit unrolled methods, while maintaining the advantage of a low GPU memory usage, as well as similar reconstruction quality to the best unrolled methods for all considered image inverse problems.

Another approach to cope with the memory issue of unrolled optimization methods is the Deep equilibrium (DEQ) model which extends unrolled methods to have a theoretically infinite amount of iterations by leveraging fixed-point properties. This allows for simpler back-propagation, which can even be done in a Jacobian-free manner. The resulting back-propagation has a memory-footprint independent of the number of iterations used. Deep equilibrium models, indeed avoid large memory consumption by back-propagating through the architecture’s fixed point. This results in an architecture that is intended to converge and allows for the prior to be more general. We have leveraged deep equilibrium models to train a plug-and-play style prior that can be used to solve inverse problems for a range of tasks.

8.4.1 Zero-error interactive compression

Participants: Nicolas Charpenay, Aline Roumy.

Zero-error source coding is at the core of practical compression algorithms. Here, zero error refers to the case where the data is decompressed without any error for any block-length. This must be differentiated from the vanishing error setup, where the probability of error of the decompressed data decreases to zero as the length of the processed data tends to infinity. In point-to point communication (one sender/one receiver), zero-error and vanishing-error information theoretic compression bounds coincides. This is however not the case in networks, with possibly many senders and/or receivers. For instance, in the source coding problem with a helper at the decoder, the compression bounds under both hypotheses do no coincide. More precisely, there exist examples, where both bounds differ, but more importantly, the zero-error compression characterization for the general case is still an open problem. Here, we consider the problem of compression with interaction, where the user can partly extract and decode what is needed for his navigation in the data. This problem can be cast into a source coding where the side information may be absent at the decoder. For this problem, we show that zero-error and vanishing error rate coincide 2231.

8.4.2 Coding for computing

Participants: Nicolas Charpenay, Aline Roumy.

Due to the ever-growing amount of visual data produced and stored, there are now evidences that these data will not only be viewed by humans but also processed by machines. In this context, practical implementations aim to provide better compression efficiency when the primary consumers are machines that perform certain computer vision tasks. One difficulty arises from the fact that these tasks are not known upon compression. Therefore we study the problem of coding for computing, when only the class of the tasks, and not the task itself is known upon compression. When the classes of tasks are quite representative (in the database, each class is requested at least twice), we derive a single letter characterization of the optimal compression bounds. More importantly, we show that there exists a coding strategy that outperforms the compression of the whole data 33. In coding for computing, the optimal coding performance depends on a characteristic graph, that represents the joint distribution of the data and the tasks. We provide new properties of the entropy of AND and OR products of these graphs. This allows to increase the number of graphs for which these graph entropies can be computed.

8.4.3 Semantic Alignment for Multi-item Compression

Participants: Tom Bachard, Thomas Maugey, Anju Jose Tom.

Coding algorithms usually compress independently the images of a collection, in particular when the correlation between them only resides at the semantic level, i.e., information related to the high-level image content. However, in many image sets, even though pixel correlation is absent (i.e., no correlation exists at the spatial level), semantic correlation can be present (i.e., high level content may be similar among the images). This should be exploited to reduce the storage burden. However, no such solution exist. Even worst, the conventional (learning-based or not) image compression of image data set leads to coded representations that are quite spread in the latent space, and thus strongly uncorrelated.

In this work, we have proposed a coding solution able to exploit the semantic redundancy to decrease the storage cost of data collections. First we formally introduced the multi-item compression framework, as the compression of a given image set, in which each image belongs to one class of object (e.g., cat, tree, train, etc.). In other words, the semantic correlation is present as several images may contain the same type of object (even though their pixel similarity is different). Then we derived a loss term to shape the latent space of a variational auto-encoder so that the latent vectors of semantically identical images (i.e., belonging to the same class) can be aligned (see Figure 5). The fact that these latent vectors are aligned further enables an inter-item coder to exploit this redundancy and reduce the bitrate. Finally, we experimentally demonstrated that this alignment leads to a more compact representation of the data collection. This work has been published in 21.

Representation of the multi-image coder latent space before (left) and after (right) semantic alignment. The numbers represent the image class.

CIFRE contract with Ateme on neural networks for video compression

Participants: Christine Guillemot, Yiqun Liu, Aline Roumy.

• Title  : Neural networks for video compression of reduced complexity
• Partners : Ateme (T. Guionnet, M. Abdoli), Inria-Rennes.
• Funding : Ateme, ANRT.
• Period : Aug.2020-Jul.2023.

The goal of this Cifre contract is to investigate deep learning architectures for the inference of coding modes in video compression algorithms with the ultimate goal of reducing the encoder complexity. The first step addresses the problem of Intra coding modes and quad-tree partitioning inference. The next step will consider Inter coding modes taking into account motion and temporal information.

Contract LITCHIE with Airbus on deep learning for satellite imaging

Participants: Denis Bacchus, Christine Guillemot, Arthur Lecert, Aline Roumy.

• Title  : Deep learning methods for low light vision
• Partners : Airbus (R. Fraisse), Inria-Rennes.
• Funding : BPI.
• Period : Sept.2020-Aug.2023.

The goal of this contract is to investigate deep learning methods for low light vision with sattelite imaging. The SIROCCO team focuses on two complementary problems: compression of low light images and restoration under conditions of low illumination, and hazing. The problem of low light image enhancement implies handling various factors simultaneously including brightness, contrast, artifacts and noise. We investigate solutions coupling the retinex theory, assuming that observed images can be decomposed into reflectance and illumination, with machine learning methods. We address the compression problem taking into account the processing tasks considered on the ground such as the restoration task, leading to an end-to-end optimization approach.

Cifre contract with MediaKind on Multiple profile encoding optimization

Participants: Reda Kaafarani, Thomas Maugey, Aline Roumy.

• Title  : Multiple profile encoding optimization
• Partners : MediaKind, Inria-Rennes.
• Funding : MediaKind, ANRT.
• Period : April 2021-April 2024.

The goal of this Cifre contract is to optimize a streaming solution taking into the whole process, namely the encoding, the long-term and the short term storages (in particular for replay, taking into the popularity of the videos), the multiple copies of a video (to adapt to both the resolution and the bandwidth of the user), and the transmissions (between all entities: encoder, back-end and front-end server, and the user). This optimization will be with several objectives as well. In particular, the goals will be to maximize the user experience but also to save energy and/or the deployment cost of a streaming solution.

10.2.1 H2020 projects

10.3.1 Project Action Exploratoire "Data Repurposing"

Participants: Anju Jose Tom, Thomas Maugey.

• Funding: Inria.
• Period: Sept. 2020 - Aug. 2023.

Lossy compression algorithms trade bits for quality, aiming at reducing as much as possible the bitrate needed to represent the original source (or set of sources), while preserving the source quality. In the exploratory action "DARE", we propose a novel paradigm of compression algorithms, aimed at minimizing the information loss perceived by the final user instead of the actual source quality loss, under compression rate constraints. In particular, we plan to measure the amount of information spanned by a data collection in the semantic domain. First, it enables to identify the high-level information contained in each of the image/video of a data collection. Second, it permits to take into account the redundancies and dissmilarities in the calculation of the global volume of information that is contained in a data collection. Finally, we propose to take into account the user's preferences in this calculation, since two users may have different tastes and priorities. Once the measure of information is set, we plan to build efficient sampling algorithms to reduce the data collection's size. This project also enables to explore new ideas for image generative compression, when part of the content can be “invented” at the decoder side.

10.3.2 IA Chair: DeepCIM- Deep learning for computational imaging with emerging image modalities

Participants: Rita Fermanian, Christine Guillemot, Brandon Lebon, Guillaume Le Guludec, Mikael Le Pendu, Jinglei Shi.

• Funding: ANR (Agence Nationale de la Recherche).
• Period: Sept. 2020 - Aug. 2024.

The project aims at leveraging recent advances in three fields: image processing, computer vision and machine (deep) learning. It will focus on the design of models and algorithms for data dimensionality reduction and inverse problems with emerging image modalities. The first research challenge will concern the design of learning methods for data representation and dimensionality reduction. These methods encompass the learning of sparse and low rank models, of signal priors or representations in latent spaces of reduced dimensions. This also includes the learning of efficient and, if possible, lightweight architectures for data recovery from the representations of reduced dimension. Modeling joint distributions of pixels constituting a natural image is also a fundamental requirement for a variety of processing tasks. This is one of the major challenges in generative image modeling, field conquered in recent years by deep learning. Based on the above models, our goal is also to develop algorithms for solving a number of inverse problems with novel imaging modalities. Solving inverse problems to retrieve a good representation of the scene from the captured data requires prior knowledge on the structure of the image space. Deep learning based techniques designed to learn signal priors, tcan be used as regularization models.

10.3.3 CominLabs MOVE project: Mature Omnidirectional Video Exploration.

Participants: Sébastien Bellenous, Navid Mahmoudian Bidgoli, Thomas Maugey, Aline Roumy.

• Funding: Labex CominLabs.
• Period: Jan. 2021 - Dec. 2022.

This project aims to secure the industrial impact of the InterCom project, ended in Dec. 2020, and funded by the Labex CominLabs. Indeed, the goal of the Cominlabs InterCom project was to design novel compression algorithms to allow interactive communication between users and a server. One instance of this interactive scenario is a visual immersive experience, where a user can navigate freely in a 3D scene. This leads to tremendous amount of visual data, such that the whole scene cannot be sent to the user. Fortunately, the user watches only a part of the scene, such that only the requested part needs to be sent to the user. This however presents a great challenge from the point of view of the compression. Indeed, the data needs to be compressed once online, but decompressed in many manners, one manner being one requested point of view. In the case of a static 360 camera, the navigation has 3 Degrees of Freedom such that there is one decompression per value of a 3D-vector. One of the achievement of the InterCom project is a prototype for an interactive compression scheme for 360-degree images. The MOVE project helps the ADT ICOV project towards the construction of a full demonstrator. It also helps the maturation of the startup project lead by Navid Mahmoudian Bidgoli (ANAX).

10.3.4 CominLabs Colearn project: Coding for Learning

Participants: Thomas Maugey, Rémi Piau, Aline Roumy.

• Partners: Inria-Rennes (Sirocco team); LabSTICC, IMT Atlantique, (team Code and SI); IETR, INSA Rennes (Syscom team).
• Funding: Labex CominLabs.
• Period: Sept. 2021 - Dec. 2024.

The amount of data available online is growing so fast that it is essential to rely on advanced Machine Learning techniques so as to automatically analyze, sort, and organize the content uploaded by e.g. sensors or users. The conventional data transmission framework assumes that the data should be completely reconstructed, even with some distortions, by the server. Instead, this project aims to develop a novel communication framework in which the server may also apply a learning task over the coded data. The project will therefore develop an Information Theoretic analysis so as to understand the fundamental limits of such systems, and develop novel coding techniques allowing for both learning and data reconstruction from the coded data.

10.3.5 Inria Start-up Studio: Anax

Participants: Navid Mahmoudian Bidgoli, Simon Evain, Thomas Maugey, Aline Roumy.

Two former PhD students of the team (Navid Mahmoudian Bidgoli and Simon Evain) plan to launch a startup, named Anax, on the theme of omnidirectional/image processing. They have obtained a one year grant (Sept 2021 - Aug 2022) for both of them funded by the Inria Startup Studio. Here is the description of the Anax project.

Anax aims to provide a deep tech software solution for processing 360-degree visual content with artificial intelligence (AI) specially designed for the preparation of virtual tours. Anax is aimed at various actors who wish to offer an immersive visit experience to improve their visibility, such as real estate agencies, cultural institutions, and interior designers. Anax is developing a technology that allows retrieving a faithful 3D reconstruction of a building from 360-degree images, opening up a wide range of applications based on AI image processing such as automatic recommendation of similar apartments, augmented reality, automatic inventory, etc. The envisaged solution, based on artificial intelligence, works even with consumer 360-degree capturing devices that are readily accessible to the general public.

10.3.6 ANR Young researcher grant: MAssive multimedia DAta collection REpurposing (MADARE)

Participants: Tom Bordin, Thomas Maugey.

• Funding: ANR (Agence Nationale de la Recherche).
• Period: Apr. 2022 - Oct. 2025.

Compression algorithms are nowadays overwhelmed by the tsunami of visual data created everyday. Despite a growing efficiency, they are always constrained to minimize the compression error, computed in the pixel domain. The Data Repurposing framework, proposed in the MADARE project, will tear down this barrier, by allowing the compression algorithm to “reinvent” part of the data at the decoding phase, and thus saving a lot of bit-rate by not coding it. Concretely, a data collection is only encoded to a compact description that is used to guarantee that the regenerated content is semantically coherent with the initial one.

In practice, it opens several research directions: how to organise the latent space (in which the coded descriptions lie) such that the information is efficiently and intelligibly represented ? How to regenerate a synthesized content from this compact description (based for example on guided diffusion algorithms) ? Finally, how to extend this idea to video ?

By revisiting the compression problem, the MADARE project aims gigantic compression ratios enabling, among other benefits, to reduce the impact of exploding data creation on the cloud servers’ energy consumption.

11.1.1 Scientific events: organisation

11.1.2 Journal

11.1.3 Invited talks

• T. Maugey gave a seminar at ENS Paris-Saclay on “Compression of visual data: beyond conventional approaches” May 2022.
• C. Guillemot gave a seminar on "distilled low-rank neural network optimization" at MidSweden Univ., Dec. 20, 2022.
• A. Roumy gave a seminar at ENS Rennes on “User-centric data compression and application to 360 videos", March 4, 2022.
• A. Roumy gave a seminar at ENSEEIHT Toulouse on “Neural Networks beyond the Euclidean case: omnidirectional images and application to compression", Oct. 20, 2022.

11.1.4 Leadership within the scientific community

• C. Guillemot has served as Chair of the evaluation panel of the computer vision centre of the Centers de Recerca de Catalunya (CERCA), Barcelona, Feb. 2022.
• A. Roumy is a member of the IEEE Image, Video, and Multidimensional Signal Processing Technical Committee (IVMSP TC).
• A. Roumy is a member of the Executive board of the National Research group in Image and Signal Processing (GRETSI).
• T. Maugey is a member of the European Association For Signal Processing (EURASIP) Technical Area Committee.

11.1.5 Scientific expertise

• C. Guillemot is a member of the Advisory Board of the Strategic Research Programme of the Vrije Universiteit Brussels
• C. Guillemot is member of the commission SEN-3 of the "Fond National de la recherche scientifique (FNRS), en Belgique.
• C. Guillemot has participated in the consolidator grant panel at the Swiss National Science Foundation (Dec. 2022).
• C. Guillemot has participated in the ANR committee selecting Franco-German research proposals in the field of artificial intelligence.
• A. Roumy has participated in the committee selecting research proposals funded by Innoviris (Institut bruxellois pour la Recherche et l'Innovation) Innovation and Research Institute, Bruxelles, Belgium.
• T. Maugey is a referee for the Italian National Agency for the Evaluation of the University and Research Systems (ANVUR)

11.1.6 Research administration

• C. Guillemot is member of the commission Recherche de l'Université de Rennes 1.
• C. Guillemot has been a member of the jury of the concours CRCN/ISFP of Rennes, April 2022.
• C. Guillemot has been a member of the jury for recruiting a Professor at the Univ. of Nancy, May 2022.
• A. Roumy has been a member of the Inria Delegation committee of the Rennes Bretagne Atlantique center, Febr. 2022.
• A. Roumy is a member of the Inria evaluation committee.
• A. Roumy has been a member of the jury of the concours CRCN/ISFP of Inria Sophia Antipolis Méditerranée, June 2022.
• T. Maugey served as a jury member of the PhD recruitment committee for Inria-Rennes, June 2022

11.2.1 Supervision

• Ipek Anil Atalay Appak (PhD in progress): “Wavefront coding for computational plenoptic cameras" (Started in August 2021, supervised by C. Guillemot and H. Caglayan(Tampere University)).
• Davi Rabbouni Freitas (PhD in progress): “Sensing and Reconstruction of Plenoptic Point Clouds" (started in September 2021, supervised by C.Guillemot and I. Tabus (Tampere University)).
• Denis Bacchus (PhD in progress): “Deep Learning for selective lossy image compression” (started in september 2020, supervised by A. Roumy and C. Guillemot).
• Tom Bachard (PhD in progress): “Extreme compression of image/video databases using GAN-based synthesis” (started in October 2021, supervised by T. Maugey).
• Tom Bordin (PhD in progress): “Extreme generative video compression” (Started in October 2022, supervised by T. Maugey).
• Nicolas Charpenay (PhD in progress): “Combinatorics and strategic aspects in Information Theory” (started in October 2020, supervised by M. le Treust and A. Roumy).
• Rita Fermanian (PhD in progress): “Deep Learning for inverse problems and application to omni-directional imaging” (started in October 2020, supervised by C. Guillemot and M. Le Pendu (INTERDIGITAL, Rennes)).
• Kai Gu (PhD in progress): “Spherical light field representation and reconstruction from omni-directional imagery" (Started in June 2021, supervised by C. Guillemot, T. Maugey and S. Knorr (Technische Universität Berlin)).
• Reda Kaafarani (PhD in progress): “Optimisation of a multi-profile encoding system” (started in April 2021, supervised by A. Roumy, T. Maugey, J. Le Tanou (MediaKind France) and M. Ropert (MediaKind France)).
• Brandon Le Bon (PhD in progress): “Deep learning for light field acquisition and restoration” (started in November 2020, supervised by C. Guillemot and M. Le Pendu (INTERDIGITAL, Rennes)).
• Arthur Lecert (PhD in progress): “Low-light image enhancement with learning techniques" (started in October 2020, supervised by C. Guillemot and A. Roumy).
• Yiqun Liu(PhD in progress): “Learning for new generation video coders" (started in Aug 2020, supervised by C. Guillemot, A. Roumy, and T. Guionnet (Ateme)).
• Rémi Piau(PhD in progress): “Coding for learning" (started in Oct 2021, supervised by A. Roumy, T. Maugey).
• Soheib Takhtardeshir (PhD in progress): “Representation and compression of multi-sensor video for combined human and computer vision applications" (Started in Jan 2022, supervised by C. Guillemot, M.Sjöström (Mid Sweden University) and co-supervised by R.Olsson(Mid Sweden University))
• Samuel Willingham (PhD in progress): “Learning algorithms for inverse problems with modern imaging modalities" (Started in June 2021, supervised by C. Guillemot and M.Sjöström (Mid Sweden University)).

11.2.2 Juries

• C. Guillemot was member, as chair, of the PhD jury of of F. Nasiri, Univ. Rennes 1, May 2022.
• C. Guillemot was member, as rapporteur, of the PhD jury of B. Chiche, Univ. Paris-Saclay, Oct. 2022
• A. Roumy was member, as rapporteure, of the HDR jury of A. Goupil, Univ. Reims Champagne-Ardenne, April 2022.
• A. Roumy was member, as rapporteure, of the PhD jury of M. Aklouf, Univ. Paris-Saclay, May 2022.
• A. Roumy was member, as chair, of the PhD jury of C. Bonnineau, Univ. Rennes 1, May 2022.
• A. Roumy was member of the PhD jury of L. Chetot, Univ. Lyon, July 2022.
• A. Roumy was member, as rapporteure, of the PhD jury of V. Alves De Oliveira, Univ. Toulouse, Oct. 2022.
• A. Roumy was member, as rapporteure, of the PhD jury of R. Bou Rouphael, Univ. Paris-Cergy, Dec. 2022.
• A. Roumy was member, of the PhD jury of A. Marmoret, Univ. Rennes 1, Dec. 2022.
• A. Roumy was member, as chair, of the PhD jury of B. Hamoum, Univ. Bretagne Sud, Dec. 2022.
• T. Maugey was member of the PhD jury of Sylvia Rossi, University College of London (UCL), London, UK, Jan. 2022

11.3.1 Education

• T. Maugey presented the research profession at Lycée Brequigny in the context of the national "Chiche" program (March, 2022)
• A. Roumy presented her research career at the Inria/Irisa women-only group meeting, June 2022.

11.3.2 Interventions

• T. Maugey has collaborated with the artist Mustapha Azeroual for the project aRza, exhibited at the Jeu de Paume, France, June-Sept. 2022.

International journals

• 7 articleH.Hadi Amirpour, C.Christine Guillemot, M.Mohammad Ghanbari and C.Christian Timmerer. Advanced Scalability for Light Field Image Coding.IEEE Transactions on Image ProcessingNovember 2022, 1-15
  HAL
• 8 articleR.Rita Fermanian, M.Mikael Le Pendu and C.Christine Guillemot. PnP-ReG: Learned Regularizing Gradient for Plug-and-Play Gradient Descent.SIAM Journal on Imaging SciencesNovember 2022, 1-28
  HALback to text
• 9 articleP.Patrick Garus, F.Felix Henry, J.Joël Jung, T.Thomas Maugey and C.Christine Guillemot. Immersive Video Coding: Should Geometry Information be Transmitted as Depth Maps?IEEE Transactions on Circuits and Systems for Video Technology325May 2022, 3250-3264
  HALDOI
• 10 articleX.Xiaoran Jiang, J.Jinglei Shi and C.Christine Guillemot. An Untrained Neural Network Prior for Light Field Compression.IEEE Transactions on Image ProcessingOctober 2022, 1-15
  HALback to text
• 11 articleG.Guillaume Le Guludec and C.Christine Guillemot. Deep Unrolling for Light Field Compressed Acquisition using Coded Masks.IEEE AccessApril 2022, 1-17
  HALDOIback to text
• 12 articleM.Mikael Le Pendu and C.Christine Guillemot. Preconditioned Plug-and-Play ADMM with Locally Adjustable Denoiser for Image Restoration Mikael.SIAM Journal on Imaging SciencesNovember 2022, 1-30
  HALback to text
• 13 articleN.Navid Mahmoudian Bidgoli, R.Roberto Azevedo, T.Thomas Maugey, A.Aline Roumy and P.Pascal Frossard. OSLO: On-the-Sphere Learning for Omnidirectional images and its application to 360-degree image compression.IEEE Transactions on Image Processing2022, 1-16
  HAL
• 14 articleJ.Jinglei Shi, X.Xiaoran Jiang and C.Christine Guillemot. Deep Residual Architecture Using Pixel and Feature Cues for View Synthesis and Temporal Interpolation.IEEE Transactions on Computational ImagingMarch 2022, 1-14
  HALDOI
• 15 articleN.Nikolaos Thomos, T.Thomas Maugey and L.Laura Toni. Machine Learning for Multimedia Communications.Sensors223January 2022, 1-31
  HALDOI
• 16 articleT.Tanaboon Tongbuasirilai, J.Jonas Unger, C.Christine Guillemot and E.Ehsan Miandji. A Sparse Non-parametric BRDF Model.ACM Transactions on GraphicsMay 2022, 1-18
  HALDOIback to text
• 17 articleZ.Zhaolin Xiao, J.Jinglei Shi, X.Xiaoran Jiang and C.Christine Guillemot. Axial refocusing precision model with light fields.Signal Processing: Image CommunicationMay 2022, 1-11
  HALDOI
• 18 articleM.Meng Zhang, H.Haiyan Jin, Z.Zhaolin Xiao and C.Christine Guillemot. A Light Field FDL-HCGH Feature in Scale-Disparity Space.IEEE Transactions on Image ProcessingSeptember 2022, 1-12
  HALback to text

International peer-reviewed conferences

• 19 inproceedingsH.Hadi Amirpour, C.Christine Guillemot and C.Christian Timmerer. FuRA: Fully Random Access Light Field Image Compression.European Workshop on Video and Image ProcessingProceedings EUVIP Conf.Lisbo, PortugalSeptember 2022, 1-6
  HAL
• 20 inproceedingsP.Pascal Bacchus, R.Renaud Fraisse, A.Aline Roumy and C.Christine Guillemot. Quasi Lossless Satellite Image Compression.IGARSS 2022 - International Geoscience and Remote Sensing SymposiumProceedingsKuala Lumpur, IndonesiaJuly 2022, 1-4
  HAL
• 21 inproceedingsT.Tom Bachard, A. J.Anju Jose Tom and T.Thomas Maugey. Semantic alignment for multi-item compression.ICIP 2022 - IEEE International Conference on Image ProcessingBordeaux, FranceIEEEOctober 2022, 1-5
  HALback to text
• 22 inproceedingsN.Nicolas Charpenay, M.Maël Le Treust and A.Aline Roumy. Zero-error source coding when side information may be present.IZS 2022 - International Zurich Seminar on Information and CommunicationZurich, Switzerland2022, 1-5
  HALback to text
• 23 inproceedingsP.Patrick Garus, F.Felix Henry, T.Thomas Maugey and C.Christine Guillemot. Decoder Side Multiplane Images using Geometry Assistance SEI for MPEG Immersive Video.MMSP 2022 - IEEE 24th International Workshop on Multimedia Signal ProcessingShanghai, ChinaIEEESeptember 2022, 1-6
  HAL
• 24 inproceedingsP.Patrick Garus, F.Felix Henry, T.Thomas Maugey and C.Christine Guillemot. Motion Compensation-based Low-Complexity Decoder Side Depth Estimation for MPEG Immersive Video.MMSP 2022 - IEEE 24th International Workshop on Multimedia Signal ProcessingShanghai, ChinaIEEESeptember 2022, 1-6
  HAL
• 25 inproceedingsK.Kai Gu, T.Thomas Maugey, S.Sebastian Knorr and C.Christine Guillemot. Omni-nerf: neural radiance field from 360° image captures.ICME 2022 - IEEE international conference on multimedia and expoTaipei, TaiwanIEEEJuly 2022, 1-6
  HALback to text
• 26 inproceedingsA.Arthur Lecert, R.Renaud Fraisse, A.Aline Roumy and C.Christine Guillemot. A new regularization for retinex decomposition of low-light images.IEEE International Conference on Image ProcessingProceedingsBordeaux, FranceIEEEOctober 2022, 1-4
  HAL
• 27 inproceedingsY.Yiqun Liu, M.Mohsen Abdoli, T.Thomas Guionnet, C.Christine Guillemot and A.Aline Roumy. Light-weight cnn-based vvc inter partitioning acceleration.IVMSP 2022 - IEEE 14th Image, Video, and Multidimensional Signal Processing WorkshopNafplio, GreeceIEEEJune 2022, 1-5
  HALback to text
• 28 inproceedingsY.Yiqun Liu, M.Mohsen Abdoli, T.Thomas Guionnet, C.Christine Guillemot and A.Aline Roumy. Statistical analysis of inter coding in vvc test model (VTM).ICIP 2022 - IEEE International Conference on Image ProcessingBordeaux, FranceIEEEOctober 2022, 1-4
  HAL
• 29 inproceedingsJ.Jinglei Shi, X.Xiaoran Jiang and C.Christine Guillemot. Deep Video Frame Rate Up-conversion Network using Feature-based Progressive Residue Refinement a.VISAPP 2022 - 17th International Conference on Computer Vision Theory and Applicationsonline, FranceFebruary 2022, 1-9
  HAL

National peer-reviewed Conferences

• 30 inproceedingsN.Navid Mahmoudian Bidgoli, R.Roberto Azevedo, T.Thomas Maugey, A.Aline Roumy and P.Pascal Frossard. Méthode d'apprentissage opérant sur la sphère: Application à la compression d'images omnidirectionnelles.GRETSI 2022 - XXVIIIème Colloque Francophone de Traitement du Signal et des ImagesNancy, FranceSeptember 2022, 1-4
  HAL

Conferences without proceedings

• 31 inproceedingsN.Nicolas Charpenay, M.Maël Le Treust and A.Aline Roumy. Codage de source zéro-erreur pour un réseau de Gray-Wyner lorsque l'information adjacente peut être présente.XXVIIIème Colloque Francophone de Traitement du Signal et des Images (GRETSI'22)Nancy, FranceSeptember 2022, 1-4
  HALback to text

Doctoral dissertations and habilitation theses

• 32 thesisT.Thomas Maugey. Visual data compression: beyond conventional approaches.Université de Rennes 1June 2022
  HAL

Reports & preprints

• 33 miscN.Nicolas Charpenay, M.Maël Le Treust and A.Aline Roumy. Zero-Error Coding for Computing with Encoder Side-Information.December 2022
  HALback to text