François Rocca

3D Head Pose Estimation for TV Setups

In this paper, we present an architecture of a system which aims to personalize the TV content to the viewer reactions. The focus of the paper is on a subset of this system which identifies moments of attentive focus in a non-invasive and continuous way. The attentive focus is used to dynamically improve the user profile by detecting which displayed media or links have drawn the user attention. Our method is based on the detection and estimation of face pose in 3D using a consumer depth camera. Two preliminary experiments were carried out to test the method and to show its link to viewer interest. This study is realized in the scenario of a TV with a second screen interaction (tablet, smartphone), a behaviour that has become common for spectators.

Robotic-assisted serious game for motor and cognitive post-stroke rehabilitation

Stroke is a major cause of long-term disability that can cause motor and cognitive impairments. New technologies such as robotic devices and serious games are increasingly being developed to improve post-stroke rehabilitation. The aim of the present project was to develop a ROBiGAME serious game to simultaneously improve motor and cognitive deficits (in particular hemiparesis and hemineglect). In this context, the difficulty level of the game was adapted to each patients performance, and this individualized adaptation was addressed as the main challenge of the game development. The game was implemented on the REAplan end-effector rehabilitation robot, which was used in continuous interaction with the game. A preliminary feasibility study of a target pointing game was run in order to validate the game features and parameters. Results showed that the game was perceived as enjoyable, and that patients reported a desire to play the game again. Most of the targets included in the game design were realistic, and they were well perceived by the patients. Results also suggested that the cognitive help strategy could include one visual prompting cue, possibly combined with an auditory cue. It was observed that the motor assistance provided by the robot was well adapted for each patients impairments, but the study results led to a suggestion that the triggering conditions should be reviewed. Patients and therapists reported the desire to receive more feedback on the patients performances. Nevertheless, more patients and therapists are needed to play the game in order to give further and more comprehensive feedback that will allow for improvements of the serious game. Future steps also include the validation of the motivation assessment module that is currently under development.

Head Pose Estimation by Perspective-n-Point Solution Based on 2D Markerless Face Tracking

In this paper, we present an optimized implementation of automatic head direction extraction for a person placed in front of his webcam. The aim is to compute the different rotation angles of the head with a non-invasive and continuous tracking. Our method is based on 2D features tracking of the face with a low cost webcam. This information is associated to a set of points from a 3D head model by perspective-n-point solution to obtain pitch, roll and yaw. These results are then compared with a reference acquired with faceLAB, a robust markerless head tracker and eye tracking system.

KINterestTV - Towards Non–invasive Measure of User Interest While Watching TV

Is it possible to determine only by observing the behavior of a user what are his interests for a media? The aim of this project is to develop an application that can detect whether or not a user is viewing a content on the TV and use this information to build the user profile and to make it evolve dynamically. Our approach is based on the use of a 3D sensor to study the movements of a user’s head to make an implicit analysis of his behavior. This behavior is synchronized with the TV content (media fragments) and other user interactions (clicks, gestural interaction) to further infer viewer’s interest. Our approach is tested during an experiment simulating the attention changes of a user in a scenario involving second screen (tablet) interaction, a behavior that has become common for spectators and a typical source of attention switches.

Tangible needle, digital haystack: tangible interfaces for reusing media content organized by similarity

This paper presents the design process of a desk-set tangible user interface for the navigation and manipulation of media content organized by content-based similarity with off-the-shelf/flea market devices. For intra-media navigation, a refurbished portable vinyl player has its inside mechanics replaced by a webcam monitoring circular gray code analyzed through computer vision for position/speed tracking. For inter-media navigation, a 3D force-feedback controller is mounted in upright position on a truss with cell clamps, repurposed as trackpad. For media recomposition, motorized faders recall the effect presets of the closest/last selected media item.

Proxemics Measurement During Social Anxiety Disorder Therapy Using a RGBD Sensors Network

In this paper, we focus on the development of a new ecological methodology to study proxemics behaviors. We based our approach on a network of RGBD cameras, calibrated together. The use of this type of sensors lets us build a 3D multiview recording installation working in various natural settings. The skeleton tracking functionalities, provided by the multiple 3D data, are a useful tool to make proxemics observation and automatically code these non-verbal cues. Our goal is to propose a new approach to study proxemics behaviors of patients suffering from social anxiety disorder to improve observation capabilities of the therapist with an unobtrusive, ecological and precise measurement system.