Search results for "multi-modal"
showing 10 items of 17 documents
Depth Attention for Scene Understanding
2022
Deep learning models can nowadays teach a machine to realize a number of tasks, even with better precision than human beings. Among all the modules of an intelligent machine, perception is the most essential part without which all other action modules have difficulties in safely and precisely realizing the target task under complex scenes. Conventional perception systems are based on RGB images which provide rich texture information about the 3D scene. However, the quality of RGB images highly depends on environmental factors, which further influence the performance of deep learning models. Therefore, in this thesis, we aim to improve the performance and robustness of RGB models with comple…
Simplification Of Painting Images For Tactile Perception By Visually Impaired Persons
2018
The access to artworks by visually impaired people requires a simplified tactile representation of paintings. This paper presents the difficulties of direct transcription of artworks and the test results of simplification of the paintings done by Australian Aborigines which don't have purely visual elements such as shadows or perspective. The implemented methodology is bottom-up: it starts with tactile representation of basic elements relevant to the understanding of the whole painting, then their association into more complex concepts. The context of associations is explained through audio-description. The results of the tests with visually impaired persons are analyzed and explained.
Recommandation de parcours de formation dans un contexte mobile
2013
National audience; Les récentes avancées dans les technologies de l'information et de la communication ont vu naître de nouvelles formes d'enseignement. L'apprentissage à distance classique s'enrichit et se transforme pour donner jour à un apprentissage plus flexible, accessible sur de multiples supports, à toute heure et en tout lieu : l'apprentissage mobile. Nos travaux portent sur la conception d'un système de recommandation basé sur le contenu, modélisé en utilisant les technologies du web sémantique. La recommandation prendra en compte l'objectif de formation, mais également les supports disponibles pour dispenser cette formation, les préférences personnelles de l'apprenant, ou encore …
Enabling Technologies on Hybrid Camera Networks for Behavioral Analysis of Unattended Indoor Environments and Their Surroundings
2008
This paper presents a layered network architecture and the enabling technologies for accomplishing vision-based behavioral analysis of unattended environments. Specifically the vision network covers both the attended environment and its surroundings by means of multi-modal cameras. The layer overlooking at the surroundings is laid outdoor and tracks people, monitoring entrance/exit points. It recovers the geometry of the site under surveillance and communicates people positions to a higher level layer. The layer monitoring the unattended environment undertakes similar goals, with the addition of maintaining a global mosaic of the observed scene for further understanding. Moreover, it merges …
An optimized algorithm of image stitching in the case of a multi-modal probe for monitoring the evolution of scars
2013
International audience; We propose a new system that makes possible to monitor the evolution of scars after the excision of a tumorous dermatosis. The hardware part of this system is composed of a new optical innovative probe with which two types of images can be acquired simultaneously: an anatomic image acquired under a white light and a functional one based on autofluorescence from the protoporphyrin within the cancer cells. For technical reasons related to the maximum size of the area covered by the probe, acquired images are too small to cover the whole scar. That is why a sequence of overlapping images is taken in order to cover the required area. The main goal of this paper is to des…
Experiences from a wearable-mobile acquisition system for ambulatory assessment of diet and activity
2017
Public health trends are currently monitored and diagnosed based on large studies that often rely on pen-and-paper data methods that tend to require a large collection campaign. With the pervasiveness of smart-phones and -watches throughout the general population, we argue in this paper that such devices and their built-in sensors can be used to capture such data more accurately with less of an effort. We present a system that targets a pan-European and harmonised architecture, using smartphones and wrist-worn activity loggers to enable the collection of data to estimate sedentary behavior and physical activity, plus the consumption of sugar-sweetened beverages. We report on a unified pilot…
Powder metallurgy processing and deformation characteristics of bulk multimodal nickel
2014
cited By 7; International audience; Spark plasma sintering was used to process bulk nickel samples from a blend of three powder types. The resulting multimodal microstructure was made of coarse (average size ∼ 135 μm) spherical microcrystalline entities (the core) surrounded by a fine-grained matrix (average grain size ∼ 1.5 μm) or a thick rim (the shell) distinguishable from the matrix. Tensile tests revealed yield strength of ∼ 470 MPa that was accompanied by limited ductility (∼ 2.8% plastic strain). Microstructure observation after testing showed debonding at interfaces between the matrix and the coarse entities, but in many instances, shallow dimples within the rim were observed indica…
Analyse et fusion d’images multimodales pour la navigation autonome
2021
Robust semantic scene understanding is challenging due to complex object types, as well as environmental changes caused by varying illumination and weather conditions. This thesis studies the problem of deep semantic segmentation with multimodal image inputs. Multimodal images captured from various sensory modalities provide complementary information for complete scene understanding. We provided effective solutions for fully-supervised multimodal image segmentation and few-shot semantic segmentation of the outdoor road scene. Regarding the former case, we proposed a multi-level fusion network to integrate RGB and polarimetric images. A central fusion framework was also introduced to adaptiv…
Multi-modal biometric authentication systems
2010
The main goal of a biometric system is to discriminate automatically subjects in a reliable and dependable way, accordingly to a specific target application. The discrimination is based on one or more types of information derived from physical or behavioural traits, such as fingerprint, face, iris, voice, hand, or signature. Applications of biometrics range from homeland security and border control to e-commerce and e-banking, including secure networking and authentication. Traditionally, biometric systems working on a single biometric feature, have many limitations, such as, trouble with data sensors, where captured sensor data are often affected by noise, distinctiveness ability, because …
An Automatic Sleep Scoring Toolbox : Multi-modality of Polysomnography Signals’ Processing
2019
Sleep scoring is a fundamental but time-consuming process in any sleep laboratory. To speed up the process of sleep scoring without compromising accuracy, this paper develops an automatic sleep scoring toolbox with the capability of multi-signal processing. It allows the user to choose signal types and the number of target classes. Then, an automatic process containing signal pre-processing, feature extraction, classifier training (or prediction) and result correction will be performed. Finally, the application interface displays predicted sleep structure, related sleep parameters and the sleep quality index for reference. To improve the identification accuracy of minority stages, a layer-w…