Section: New Results

Introduction

This year Stars has proposed new results related to its three main research axes : perception for activity recognition, semantic activity recognition and software engineering for activity recognition.

Perception for Activity Recognition

Participants : François Brémond, Etienne Corvée, Antitza Dancheva, Furqan Muhammad Khan, Michal Koperski, Thi Lan Anh Nguyen, Javier Ortiz, Remi Trichet, Ujjwal Ujjwal, Srijan Das, Monique Thonnat.

The new results for perception for activity recognition are:

• Pedestrian detection: Training set optimization (see 7.2)

• Pedestrian Detection Using Deep Learning (see 7.3)

• Deep Learning applied on Embedded Systems for people detection (see 7.4)

• Facial Analysis (see 7.5)

• Multi-Object Tracking using Multi-Channel Part Appearance Representation (see 7.6)

• Tracklets Pre-Processing for Signature Computation in the Context of Multi-Shot Person Re-Identification (see 7.7)

• Multi-shot Person Re-identification in surveillance videos (see 7.8)

• Person Re-Identification using Pose-Driven Body Parts (see 7.9)

• Human Action Recognition in Videos with Local Representation (see 7.10)

• Action Detection in Untrimmed Videos (see 7.11)

• RGB-D based Action Recognition using CNNsf (see 7.12)

• Recognizing Human Actions Using RGB Sport Videos From the Web (see 7.13)

Semantic Activity Recognition

Participants : Carlos Fernando Crispim Junior, Kartik Kartik, Farhood Negin, Thanh Hung Nguyen, Kuan-Ru Lee, Antitza Dantcheva, Auriane Gros, Alexandra Koening, Guillaume Sacco, Philippe Robert, François Brémond, Monique Thonnat.

For this research axis, the contributions are :

• Event Recognition Based on Depth Image (see 7.14)

• Recognition of Daily Activities by Embedding Visual Features within a Semantic Language (see 7.15)

• Cognitive Assessment Using Gesture Recognition (see 7.16)

• Geometric and Visual Features Fusion for Action Recognition (see 7.17)

• Probabilistic Logic for Activity Recognition (see 7.18)

• Recognizing Retracing of Steps Using Walk Comparison (see 7.19)

• Safe & Easy Environment for Alzheimer Disease and related disorders (see 7.20)

• Early detection of cognitive disorders such as dementia on the basis of speech analysis ELEMENT (see 7.21)

• Serious Exergames for Cognitive Stimulation (see 7.22)

Software Engineering for Activity Recognition

Participants : Sabine Moisan, Annie Ressouche, Jean-Paul Rigault, Ines Sarray, Thanh Hung Nguyen, Daniel Gaffé, Julien Badie, Anais Ducoffe, Dorine Havyarimana, Cedric Girard-Riboulleau, François Brémond, Minh Khue Phan Tran, Philippe Robert.

The contributions for this research axis are:

• Defining an activity description language for end-users and its semantics (see 7.23)

• The Clem Workflow (see 7.24)

• Study of Temporal Properties of Neuronal Archetypes (see 7.25)

• Maintaining the engagement of older adults with dementia while interacting with serious game(see 7.26)

• Application of deep learning on healthcare (see 7.27)

• Brick & Mortar Cookies (see 7.28)