3.1 Shape and Appearance Modeling

Standard acquisition platforms, including commercial solutions proposed by companies such as Microsoft, 3dMD or 4DViews, now give access to precise 3D models with geometry, e.g. meshes, and appearance information, e.g. textures. Still, state-of-the-art solutions are limited in many respects: They generally consider limited contexts and close setups with typically at most a few meter side lengths. As a result, many dynamic scenes, even a body running sequence, are still challenging situations; They also seldom exploit time redundancy; Additionally, data driven strategies are yet to be fully investigated in the field. The MORPHEO team builds on the Kinovis platform for data acquisition and has addressed these issues with, in particular, contributions on time integration, in order to increase the resolution for both shapes and appearances, on representations, as well as on exploiting machine learning tools when modeling dynamic scenes. Our originality lies, for a large part, in the larger scale of the dynamic scenes we consider as well as in the time super resolution strategy we investigate. Another particularity of our research is a strong experimental foundation with the multiple camera Kinovis platforms.

3.2 Dynamic Shape Vision

Dynamic Shape Vision refers to research themes that consider the motion of dynamic shapes, with e.g. shapes in different poses, or the deformation between different shapes, with e.g. different human bodies. This includes for instance shape tracking and shape registration, which are themes covered by MORPHEO. While progress has been made over the last decade in this domain, challenges remain, in particular due to the required essential task of shape correspondence that is still difficult to perform robustly. Strategies in this domain can be roughly classified into two categories: (i) data driven approaches that learn shape spaces and estimate shapes and their variations through space parameterizations; (ii) model based approaches that use more or less constrained prior models on shape evolutions, e.g. locally rigid structures, to recover correspondences. The MORPHEO team is substantially involved in both categories. The second one leaves more flexibility for shapes that can be modeled, an important feature with the Kinovis platform, while the first one is interesting for modeling classes of shapes that are more likely to evolve in spaces with reasonable dimensions, such as faces and bodies under clothing. The originality of MORPHEO in this axis is to go beyond static shape poses and to consider also the dynamics of shape over several frames when modeling moving shapes, and in particular with shape tracking, animation and, more recently, face registration.

3.3 Inside Shape Vision

Another research axis is concerned with the ability to perceive inside moving shapes. This is a more recent research theme in the MORPHEO team that has gained importance. It was originally the research associated to the Kinovis platform installed in the Grenoble Hospitals. This platform is equipped with two X-ray cameras and ten color cameras, enabling therefore simultaneous vision of inside and outside shapes. We believe this opens a new domain of investigation at the interface between computer vision and medical imaging. Interesting issues in this domain include the links between the outside surface of a shape and its inner parts, especially with the human body. These links are likely to help understanding and modeling human motions. Until now, numerous dynamic shape models, especially in the computer graphics domain, consist of a surface, typically a mesh, rigged to a skeletal structure that is never observed in practice but allows to parameterize human motion. Learning more accurate relationships using observations can therefore significantly impact the domain.

3.4 Shape Animation

3D animation is a crucial part of digital media production with numerous applications, in particular in the game and motion picture industry. Recent evolutions in computer animation consider real videos for both the creation and the animation of characters. The advantage of this strategy is twofold: it reduces the creation cost and increases realism by considering only real data. Furthermore, it allows to create new motions, for real characters, by recombining recorded elementary movements. In addition to enable new media contents to be produced, it also allows to automatically extend moving shape datasets with fully controllable new motions. This ability appears to be of great importance with the recent advent of deep learning techniques and the associated need for large learning datasets. In this research direction, we investigate how to create new dynamic scenes using recorded events. More recently, this also includes applying machine learning to datasets of recorded human motions to learn motion spaces that allow to synthesize novel realistic animations.

4.1 4D modeling

Modeling shapes that evolve over time, analyzing and interpreting their motion has been a subject of increasing interest of many research communities including the computer vision, the computer graphics and the medical imaging communities. Recent evolutions in acquisition technologies including 3D depth cameras (Time-of-Flight and Kinect), multi-camera systems, marker based motion capture systems, ultrasound and CT scanners have made those communities consider capturing the real scene and their dynamics, create 4D spatio-temporal models, analyze and interpret them. A number of applications including dense motion capture, dynamic shape modeling and animation, temporally consistent 3D reconstruction, motion analysis and interpretation have therefore emerged.

4.2 Shape Analysis

Most existing shape analysis tools are local, in the sense that they give local insight about an object's geometry or purpose. The use of both geometry and motion cues makes it possible to recover more global information, in order to get extensive knowledge about a shape. For instance, motion can help to decompose a 3D model of a character into semantically significant parts, such as legs, arms, torso and head. Possible applications of such high-level shape understanding include accurate feature computation, comparison between models to detect defects or medical pathologies, and the design of new biometric models.

4.3 Human Motion Analysis

The recovery of dense motion information enables the combined analysis of shapes and their motions. Typical examples include the estimation of mean shapes given a set of 3D models or the identification of abnormal deformations of a shape given its typical evolutions. The interest arises in several application domains where temporal surface deformations need to be captured and analyzed. It includes human body analyses for which potential applications are anyway numerous and important, from the identification of pathologies to the design of new prostheses.

4.4 Virtual and Augmented Reality

This domain has actually seen new devices emerge that enable now full 3D visualization, for instance the HTC Vive, the Microsoft Hololens and the Magic leap one. These devices create a need for adapted animated 3D contents that can either be generated or captured. We believe that captured 4D models will gain interest in this context since they provide realistic visual information on moving shapes that tend to avoid negative perception effects such as the uncanny valley effect. Furthermore, captured 4D models can be exploited to learn motion models that allow for realistic generation, for instance in the important applicative case of generating realistically moving human avatars. Besides 3D visualization devices, many recent applications also rely on everyday devices, such as mobile phones, to display augmented reality contents with free viewpoint ability. In this case, 3D and 4D contents are also expected.

5.1 Footprint of research activities

We are concerned by the environmental impact of our research. While we have yet no tools to exactly measure the footprint of our activities, we are anyway evolving our practice on different aspects.

Dissemination strategy: Traditionally, Morpheo's dissemination strategy has been to publish in the top conferences of the field (CVPR, ECCV, ICCV, MICCAI), as well as to give invited talks at various locations all over the world, leading therefore to sometimes long disance work trips. Even if Morpheo's members have never been frequent long distance travelers their practice has changed. As a result of the Covid-19 crisis, our research community has started in 2020 to experiment virtual conferences. These digital meetings have some merits: beyond the obvious advantage of reduced travel (leading to fewer emissions and improved work-life balance), they democratize attendance as attending has a fraction of the previous cost. On the other hand, an important drawback is that digital attendance leads to low-quality exchanges between participants as on-line discussions can be often difficult, which actually tend to reduce the global attendance. Within the team we currently try to increase submissions to top journals in the field directly, hence reducing travels, while still attending some in-person conferences or seminars to allow for networking.

Data management: The data produced by the Morpheo team occupies large memory volumes. The Kinovis platform at Inria Grenoble typically produces around 1,5GB per second when capturing one actor at 50fps. The platform that also captures X-ray images at CHU produces around 1.3GB of data per second at 30fps for video and X-rays. For practical reasons, we reduce the data as much as possible with respect to the targeted application by only keeping e.g. 3D reconstructions or down-sampled spatial or temporal camera images. Yet, acquiring, processing and storing these data is costly in terms of resources. In addition, the data captured by these platforms are personal data with highly constrained regulations. Our data management therefore needs to consider multiple aspects: data encryption to protect personal information, data backup to allow for reproducibility, and the environmental impact of data storage and processing. For all these aspects, we are constantly checking for new and more satisfactory solutions.

For data processing, we rely heavily on cluster uses, in particular with neural networks which are known to have a heavy carbon footprint. Yet, in our research field, these types of processing methods have been shown to lead to significant performance gains. For this reason, we continue to use neural networks as tools, while attempting to use architectures that allow for faster and more energy efficient training, and simple test cases that can often be trained on local machines (with GPU) to allow for less costly network design. A typical example of this type of research is with the work of Briac Toussaint, whose first thesis contribution exhibits a high quality reconstruction algorithm for the Kinovis platform, with under a minute of GPU computation time per frame, an order of magnitude faster than our previous research solution, yet it achieves improved precision 10.

5.2 Impact of research results

Morpheo's main research topic is not related to sustainability. Yet, some of our research directions could be applied to solve problems relevant to sustainability such as:

• Realistic virtual avatar creation holds the potential to allow for realistic interactions through large geographic distances, leading to more realistic and satisfactory teleconferencing systems. As has been observed during the Covid-19 crisis, such systems can remove the need for traveling, thereby preventing costs and pollution and preserving resources. If avatars were to allow for more realistic interactions than existing teleconferencing systems, there is hope that the need to travel can be reduced even in times without health emergency. Morpheo captures and analyzes humans in 4D, thereby allowing to capture both their shapes and movement patterns, and actively works on human body modeling. These research directions tackle problems that are highly relevant for the creation of digital avatars that do not fall into the uncanny valley (even during movement). In this line of work, Morpheo particpated in the IPL Avatar project, in particular with the work of the Ph.D. student Jean Basset, and will follow up with the Nemo.AI joint lab with Interdigital on advancing this topic.
• Modeling and analyzing humans in clothing holds the potential to reduce the rate of returns of online sales in the clothing industry, which are currently high. In particular, highly realistic dynamic clothing models could help a consumer better evaluate the fit and dynamic aspects of the clothing. Morpheo's interest includes the analysis of clothing deformations based on 4D captures. Preliminary results have been achieved in the past (works of the former Ph.D. student Jinlong Yang), and we plan to continue investigating this line of work. To this end, our plans for future work include the capture a large-scale dataset of humans in different clothing items performing different motions in collaboration with NaverLabs Research. No such dataset exists to date, and it would allow the research community to study the dynamic effects of clothing.

Of course, as with any research direction, ours can also be used to generate technologies that are resource-hungry and whose necessity may be questionable, such as inserting animated 3D avatars in scenarios where simple voice communication would suffice, for instance.

6.1 Dagstuhl organisation

Stefanie Wuhrer co-organized Dagstuhl seminar 3D Morphable Models and Beyond, Dagstuhl, Germany, March 20 – 25, 2022.

7.1 New software

8.1 Pyramidal Signed Distance Learning for Spatio-Temporal Human Shape Completion

Participants: Boyao Zhou, Jean-Sébastien Franco, Martin Delagorce, Edmond Boyer.

In this figure, we show the shape completion ability from partial front depth scan, on an instance of the CAPE dataset.

This work addresses the problem of completing partial human shape observations as obtained with a depth camera. Existing methods that solve this problem can provide robustness, with for instance model-based strategies that rely on parametric human models, or precision, with learning approaches that can capture local geometric patterns using implicit neural representations. We investigate how to combine both properties with a novel pyramidal spatio-temporal learning model. This model exploits neural signed distance fields in a coarse-to-fine manner, in order to benefit from the ability of implicit neural representations to preserve local geometry details while enforcing more global spatial consistency for the estimated shapes through features at coarser levels. In addition, our model also leverages temporal redundancy with spatiotemporal features that integrate information over neighboring frames. Experiments on standard datasets show that both the coarse-to-fine and temporal aggregation strategies contribute to outperform the state-of-the-art methods on human shape completion. This work was published at Asian Conference on Computer Vision 11, and the software is made available to the research community (7.1.7).

8.2 Fast Gradient Descent for Surface Capture Via Differentiable Rendering

Participants: Briac Toussaint, Maxime Genisson, Jean-Sébastien Franco.

Differential rendering has recently emerged as a powerful tool for image-based rendering or geometric reconstruction from multiple views, with very high quality. Up to now, such methods have been benchmarked on generic object databases and promisingly applied to some real data, but have yet to be applied to specific applications that may benefit. In this paper, we investigate how a differential rendering system can be crafted for raw multi-camera performance capture. We address several key issues in the way of practical usability and reproducibility, such as processing speed, explainability of the model, and general output model quality. This leads us to several contributions to the differential rendering framework. In particular we show that a unified view of differential rendering and classic optimization is possible, leading to a formulation and implementation where complete non-stochastic gradient steps can be analytically computed and the full per frame data stored in video memory, yielding a straightforward and efficient implementation. We also use a sparse storage and coarse-to-fine scheme to achieve extremely high resolution with contained memory and computation time. We show experimentally that results rivaling in quality with state of the art multi-view human surface capture methods are achievable in a fraction of the time, typically around a minute per frame. This work was published at the 3D Visions conference 2022 10. The code for this work is made available to the community (7.1.4).

8.3 Temporal Shape Transfer Network for 3D Human Motion

Participants: João Regateiro, Edmond Boyer.

This work presents a learning-based approach to perform human shape transfer between an arbitrary 3D identity mesh and a temporal motion sequence of 3D meshes. Recent approaches tackle the human shape and pose transfer on a per-frame basis and do not yet consider the valuable information about the motion dynamics, e.g., body or clothing dynamics, inherently present in motion sequences. Recent datasets provide such sequences of 3D meshes, and this work investigates how to leverage the associated intrinsic temporal features in order to improve learning-based approaches on human shape transfer. These features are expected to help preserve temporal motion and identity consistency over motion sequences. To this aim, we introduce a new network architecture that takes as input successive 3D mesh frames in a motion sequence and which decoder is conditioned on the target shape identity. Training losses are designed to enforce temporal consistency between poses as well as shape preservation over the input frames. Experiments demonstrate substantially qualitative and quantitative improvements in using temporal features compared to optimization-based and recent learning-based methods. This work was published at the 3D Visions conference 2022 9.

8.4 A Structured Latent Space for Human Body Motion Generation

Participants: Mathieu Marsot, Stefanie Wuhrer, Jean-Sébastien Franco, Stéphane Durocher.

We propose a framework to learn a structured latent space to represent 4D human body motion, where each latent vector encodes a full motion of the whole 3D human shape. On one hand several data-driven skeletal animation models exist proposing motion spaces of temporally dense motion signals, but based on geometrically sparse kinematic representations. On the other hand many methods exist to build shape spaces of dense 3D geometry, but for static frames. We bring together both concepts, proposing a motion space that is dense both temporally and geometrically. Once trained, our model generates a multi-frame sequence of dense 3D meshes based on a single point in a low-dimensional latent space. This latent space is built to be structured, such that similar motions form clusters. It also embeds variations of duration in the latent vector, allowing semantically close sequences that differ only by temporal unfolding to share similar latent vectors. We demonstrate experimentally the structural properties of our latent space, and show it can be used to generate plausible interpolations between different actions. We also apply our model to 4D human motion completion, showing its promising abilities to learn spatio-temporal features of human motion. This work was published at the International Conference on 3D Vision 8. The software demonstrator for this work is made available to the community (7.1.3).

8.5 Impact of Self-Contacts on Perceived Pose Equivalences

Participants: Jean Basset, Badr Ouannas, Ludovic Hoyet, Franck Multon, Stefanie Wuhrer.

Defining equivalences between poses of different human characters is an important problem for imitation research, human pose recognition and deformation transfer. However, pose equivalence is a subjective information that depends on context and on the morphology of the characters. A common hypothesis is that interactions between body surfaces, such as self-contacts, are important attributes of human poses, and are therefore consistently included in animation approaches aiming at retargeting human motions. However, some of these self-contacts are only present because of the morphology of the character and are not important to the pose, e.g. contacts between the upper arms and the torso during a standing A-pose. In this work, we conduct a first study towards the goal of understanding the impact of self-contacts between body surfaces on perceived pose equivalences. More specifically, we focus on contacts between the arms or hands and the upper body, which are frequent in everyday human poses. We conduct a study where we present to observers two models of a character mimicking the pose of a source character, one with the same self-contacts as the source, and one with one self-contact removed, and ask observers to select which model best mimics the source pose. We show that while poses with different self-contacts are considered different by observers in most cases, this effect is stronger for self-contacts involving the hands than for those involving the arms. This work was published at the International Conference for Motion in Games 6.

8.6 A Visual Approach to Measure Cloth-Body and Cloth-Cloth Friction

Participants: Abdullah-Haroon Rasheed, Victor Romero, Florence Bertails-Descoubes, Stefanie Wuhrer, Jean-Sébastien Franco, Arnaud Lazarus.

Measuring contact friction in soft-bodies usually requires a specialised physics bench and a tedious acquisition protocol. This makes the prospect of a purely non-invasive, video-based measurement technique particularly attractive. Previous works have shown that such a video-based estimation is feasible for material parameters using deep learning, but this has never been applied to the friction estimation problem which results in even more subtle visual variations. Because acquiring a large dataset for this problem is impractical, generating it from simulation is the obvious alternative. However, this requires the use of a frictional contact simulator whose results are not only visually plausible, but physically-correct enough to match observations made at the macroscopic scale. In this paper, which is an extended version of our former work [31], we propose to our knowledge the first non-invasive measurement network and adjoining synthetic training dataset for estimating cloth friction at contact, for both cloth-hard body and cloth-cloth contacts. To this end we build a protocol for validating and calibrating a state-of-the-art frictional contact simulator, in order to produce a reliable dataset. We furthermore show that without our careful calibration procedure, the training fails to provide accurate estimation results on real data. We present extensive results on a large acquired test set of several hundred real video sequences of cloth in friction, which validates the proposed protocol and its accuracy. This result was published in IEEE Transactions on Pattern Analysis and Machine Intelligenge 5.

8.7 OSSO: Obtaining Skeletal Shape from Outside

Participants: Marilyn Keller, Silvia Zuffi, Michael Black, Sergi Pujades.

We address the problem of inferring the anatomic skeleton of a person, in an arbitrary pose, from the 3D surface of the body; i.e. we predict the inside (bones) from the outside (skin). This has many applications in medicine and biomechanics. Existing state-of-the-art biomechanical skeletons are detailed but do not easily generalize to new subjects. Additionally, computer vision and graphics methods that predict skeletons are typically heuristic, not learned from data, do not leverage the full 3D body surface, and are not validated against ground truth. To our knowledge, our system, called OSSO (Obtaining Skeletal Shape from Outside), is the first to learn the mapping from the 3D body surface to the internal skeleton from real data. We do so using 1000 male and 1000 female dual-energy X-ray absorptiometry (DXA) scans. To these, we fit a parametric 3D body shape model (STAR) to capture the body surface and a novel part-based 3D skeleton model to capture the bones. This provides inside/outside training pairs. We model the statistical variation of full skeletons using PCA in a pose-normalized space. We then train a regressor from body shape parameters to skeleton shape parameters and refine the skeleton to satisfy constraints on physical plausibility. Given an arbitrary 3D body shape and pose, OSSO predicts a realistic skeleton inside. In contrast to previous work, we evaluate the accuracy of the skeleton shape quantitatively on held out DXA scans, outperforming the state-of-the art. We also show 3D skeleton prediction from varied and challenging 3D bodies. The code to infer a skeleton from a body shape is available for research, and the dataset of paired outer surface (skin) and skeleton (bone) meshes are available as a Biobank Returned Dataset. This research has been conducted using the UK Biobank Resource. This result was published at the Computer Vision and Pattern Recognition conference 7 and the OSSO software is made available to the community (7.1.5) (Project page link).

8.8 Vertebrae localization, segmentation and identification using a graph optimization and an anatomic consistency cycle

Participants: Di Meng, Eslam Mohammed, Edmond Boyer, Sergi Pujades.

Vertebrae localization, segmentation and identification in CT images is key to numerous clinical applications. While deep learning strategies have brought to this field significant improvements over recent years, transitional and pathological vertebrae are still plaguing most existing approaches as a consequence of their poor representation in training datasets. Alternatively, proposed non-learning based methods take benefit of prior knowledge to handle such particular cases. In this work we propose to combine both strategies. To this purpose we introduce an iterative cycle in which individual vertebrae are recursively localized, segmented and identified using deep networks, while anatomic consistency is enforced using statistical priors. In this strategy, the transitional vertebrae identification is handled by encoding their configurations in a graphical model that aggregates local deep-network predictions into an anatomically consistent final result. Our approach achieves the state-of-the-art results on the VerSe20 challenge benchmark, and outperforms all methods on transitional vertebrae as well as the generalization to the VerSe19 challenge benchmark. Furthermore, our method can detect and report inconsistent spine regions that do not satisfy the anatomic consistency priors. The code and model are available for research purposes (7.1.6). This result was published at the Computer Vision and Pattern Recognition conference 14.

9.1 Bilateral contracts with industry

Participants: Boyao Zhou, Pierre Zins, Stefanie Wuhrer, Jean-Sébastien Franco, Edmond Boyer.

1. The Morpheo INRIA team and Microsoft research set up a collaboration on the capture and modelling of moving shapes using multiple videos. Two PhD works are part of this collaboration with the objective to make contributions on 4D Modeling. The PhDs take place at Inria Grenoble Rhône-Alpes and involve visits and stays at Microsoft in Cambridge (UK) and Zurich (CH). The collaboration is part of the Microsoft Research - Inria Joint Centre. One of the PhD (Matthieu Armendo) has finished in 2021 and the other PhD (Boyao Zhou) is pursuing has finished in November 2022.
2. The Morpheo INRIA team has another collaboration with Meta reality lab in San Francisco. The collaboration involves one PhD who is currently at the INRIA Grenoble Rhône-Alpes working on the estimation of shape and appearance from a single image. The collaboration started in 2019 and is ongoing with the PhD of Pierre Zins.
3. The Morpheo INRIA team has a collaboration with Interdigital in Rennes through the Nemo.AI joint lab. The kickoff to this collaboration has happened in November 2022. The collaboration will involve two PhD co-supervisions starting in 2023, one in INRIA Rennes and one at the INRIA de l'Université Grenoble Alpes. The subject of the collaboration revolves around digital humans, one the one hand to estimate the clothing of humans from images, and on the other to estimate hair models from one or several videos.

9.2 Bilateral grants with industry

Participants: Sergi Pujades Rocamora, Edmond Boyer, Jean-Sébastien Franco.

1. The Morpheo INRIA team collaborates with the local start-up Anatoscope created by former researchers at INRIA Grenoble. The collaboration involves one CIFRE PhD (Nicolas Comte) who shares his time between Morpheo and Anatoscope offices. He is working on new classifications methods of the scoliose disease that take the motion of the patient into account in addition to their morphology. The collaboration started in 2020.
2. The Morpheo team is also involved in a CIFRE PhD co-supervision (Anilkumar Swamy) with Naver Labs since July 2021, with Gregory Rogez, Mathieu Armando and Vincent Leroy. The work revolves around monocular hand-object reconstruction by showing the object to the camera.

10.1 International initiatives

10.2 International research visitors

10.3 National initiatives

11.1 Promoting scientific activities

• Edmond Boyer and Sergi Pujades co-organized the MICCAI Challenge 3DTeeth, held in Singapour during the MICCAI conference. (3DTeeth) September 18, 2022.

11.2 Teaching - Supervision - Juries

12.1 Major publications

• 1 inproceedingsM.Marilyn Keller, S.Silvia Zuffi, M.Michael Black and S.Sergi Pujades. OSSO: Obtaining Skeletal Shape from Outside.2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)New Orleans, United StatesIEEEJune 2022, 20460-20469
  HALDOI
• 2 inproceedingsM.Mathieu Marsot, S.Stefanie Wuhrer, J.-S.Jean-Sébastien Franco and S.Stephane Durocher. A Structured Latent Space for Human Body Motion Generation.International Conference on 3D Vision (3DV)Prague, Czech RepublicSeptember 2022
  HAL
• 3 inproceedingsB.Briac Toussaint, M.Maxime Genisson and J.-S.Jean-Sébastien Franco. Fast Gradient Descent for Surface Capture Via Differentiable Rendering.3DV 2022 - International Conference on 3D VisionPrague, Czech RepublicSeptember 2022, 1-10
  HAL

12.2 Publications of the year