SciTePress - Affiliation Profile

INRIA, France

Context (please use complete words only)

Field

Sort by

18 Papers

TrichANet: An Attentive Network for Trichogramma Classification

Topics: Visual Attention and Image Saliency; Assistive Computer Vision

Authors: Agniv Chatterjee ¹ , Snehashis Majhi ¹ , Vincent Calcagno ² and François Brémond ¹

Affiliations: ¹ INRIA Sophia Antipolis, 2004 Route des Lucioles, 06902, Valbonne, France , ² INRAE, Sophia Antipolis FR, Rte des Chappes, 06560, France

Digital Object Identifier: 10.5220/0011677700003417

Proceedings of VISAPP 2023/VISIGRAPP, ISBN: 978-989-758-634-7

Quick Abstract

Trichogramma wasp classification has a significant application in agricultural research, thanks to their massive usage and production in cropping as a bio-control agent. However, classifying these tiny species is a challenging task due to two factors: (i) Detection of these tiny wasps (barely visible with the naked eyes), (ii) Less inter-species discriminative visual features. To combat this, we propose a robust method to detect and classify the wasps from high-resolution images. The proposed method is enabled by a trich detection module that can be plugged into any competitive object detector for improved wasp detection. Further, we propose a multi-scale attention block to encode the inter-species discriminative representation by exploiting the coarse and fine-level morphological structure of the wasps for enhanced wasps classification. The proposed method along with its two key modules is validated in an in-house Trich dataset and a classification performance gain of 4% compared to recently reported baseline approaches outlines the robustness of our method. The code is available at https://github.com/ac5113/TrichANet.

DeTracker: A Joint Detection and Tracking Framework

Topics: Machine Learning Technologies for Vision; Features Extraction; Object Detection and Localization; Deep Learning for Tracking; 3D Deep Learning

Authors: Juan Diego Gonzales Zuniga , Ujjwal and François Bremond

Affiliation: INRIA Sophia Antipolis, 2004 Route des Lucioles, BP93 Sophia Antipolis Cedex, 06902, France

Digital Object Identifier: 10.5220/0010875700003124

Proceedings of VISAPP 2022/VISIGRAPP, ISBN: 978-989-758-555-5

Quick Abstract

We propose a unified network for simultaneous detection and tracking. Instead of basing the tracking framework on object detections, we focus our work directly on tracklet detection whilst obtaining object detection. We take advantage of the spatio-temporal information and features from 3D CNN networks and output a series of bounding boxes and their corresponding identifiers with the use of Graph Convolution Neural Networks. We put forward our approach in contrast to traditional tracking-by-detection methods, the major advantages of our formulation are the creation of more reliable tracklets, the enforcement of the temporal consistency, and the absence of data association mechanism for a given set of frames. We introduce DeTracker, a truly joint detection and tracking network. We enforce an intra-batch temporal consistency of features by enforcing a triplet loss over our tracklets, guiding the features of tracklets with different identities separately clustered in the feature space. Our approach is demonstrated on two different datasets, including natural images and synthetic images, and we obtain 58.7% on MOT and 56.79% on a subset of the JTA-dataset.

Video-based Behavior Understanding of Children for Objective Diagnosis of Autism

Topics: Event and Human Activity Recognition; Medical Image Applications; Deep Learning for Visual Understanding ; Assistive Computer Vision; Transfer Learning

Authors: Abid Ali ^{1

;

2} , Farhood Negin ² , Susanne Thümmler ^{1

;

2} and Francois Bremond ^{1

;

2}

Affiliations: ¹ INRIA, France , ² Université Cote d’Azur, France

Digital Object Identifier: 10.5220/0010839200003124

Proceedings of VISAPP 2022/VISIGRAPP, ISBN: 978-989-758-555-5

Quick Abstract

One of the major diagnostic criteria for Autism Spectrum Disorder (ASD) is the recognition of stereotyped behaviors. However, it primarily relies on parental interviews and clinical observations, which result in a prolonged diagnosis cycle preventing ASD children from timely treatment. To help clinicians speed up the diagnosis process, we propose a computer-vision-based solution. First, we collected and annotated a novel dataset for action recognition tasks in videos of children with ASD in an uncontrolled environment. Second, we propose a multi-modality fusion network based on 3D CNNs. In the first stage of our method, we pre- process the RGB videos to get the ROI (child) using Yolov5 and DeepSORT algorithms. For optical flow extraction, we use the RAFT algorithm. In the second stage, we perform extensive experiments on different deep learning frameworks to propose a baseline. In the last stage, a multi-modality-based late fusion network is proposed to classify and evaluate performance of ASD children. The results revealed that the multi-modality fusion network achieves the best accuracy as compared to other methods. The baseline results also demonstrate the potential of an action-recognition-based system to assist clinicians in a reliable, accurate, and timely diagnosis of ASD disorder.

Beyond Classical SERVICE Clause in Federated SPARQL Queries: Leveraging the Full Potential of URI Parameters

Topics: Semantic Interoperability; Web based Integration Technologies, Web Services, REST and CRUD services; Linked Data, Big Data and Applications in Companies

Authors: Olivier Corby , Catherine Faron , Fabien Gandon , Damien Graux and Franck Michel

Affiliation: Université Côte d’Azur, Inria, CNRS, I3S, Sophia Antipolis, France

Digital Object Identifier: 10.5220/0010660300003058

Proceedings of WEBIST 2021, ISBN: 978-989-758-536-4

Quick Abstract

Semantic Web applications integrating very different software and data sources have to face the heterogeneity of the quality and compliance to standards exhibited by each involved resource. In this paper we propose a uniform way of adapting and customizing the behavior of both the client and the server components of an HTTP exchange to cope with this diversity. We revisit the classical SERVICE clause in SPARQL federated queries in order to parameterize the behavior of both the SPARQL client and the SPARQL service. We propose mechanisms to identify and specify SPARQL federated query services and extended SPARQL-based services.

From Linked Data Querying to Visual Search: Towards a Visualization Pipeline for LOD Exploration

Topic: Web Interfaces and Applications

Authors: Aline Menin ¹ , Catherine Faron ¹ , Olivier Corby ¹ , Carla Dal Sasso Freitas ² , Fabien Gandon ¹ and Marco Winckler ¹

Affiliations: ¹ Univ. Côte d’Azur, CNRS, Inria, Sophia Antipolis, France , ² Institute of Informatics, Federal University of Rio Grande do Sul, Porto Alegre, Brazil

Digital Object Identifier: 10.5220/0010654600003058

Proceedings of WEBIST 2021, ISBN: 978-989-758-536-4

Quick Abstract

Over the recent years, Linked Open Data (LOD) has been increasingly used to support decision-making processes in various application domains. For that purpose, an increasing interest in information visualization has been observed in the literature as a suitable solution to communicate the knowledge described in LOD data sources. Nonetheless, transforming raw LOD data into a graphical representation (the so-called visualization pipeline) is not a straightforward process and often requires a set of operations to transform data into meaningful visualizations that suit users’ needs. In this paper, we propose a LOD generic visualization pipeline and discuss the implications of the internal operations (import → transform → map → render → interact) for creating meaningful visualizations of LOD datasets. To demonstrate the feasibility of this generic visualization pipeline, we implement it as the tool LDViz (Linked Data Visualizer). We demonstrate how LDViz supports access to any SPARQL endpoint through multiple use cases, allowing the users to perform searches with SPARQL queries and visualize the results using multiple visualization techniques.

STTL - A SPARQL-based Transformation Language for RDF

Topics: XML and Data Management; Metadata and Metamodeling; Ontology and the Semantic Web

Authors: Olivier Corby ¹ and Catherine Faron-Zucker ²

Affiliations: ¹ Inria Sophia Antipolis Méditerranée, France , ² Univ. Nice Sophia Antipolis, France

Digital Object Identifier: 10.5220/0005450604660476

Proceedings of WEBIST 2015, ISBN: 978-989-758-106-9

Quick Abstract

The general research question addressed in this paper is How to transform RDF into other languages. This is of prime interest to present data selected and extracted from the Web of data in a format suitable for the user (e.g., HTML or CSV). Moreover, RDF can be viewed as a meta-model to represent on the Web of data other languages and models. The above research question then becomes How to generate the concrete syntax of expressions of a given language from their RDF representation. To answer these questions, we present SPARQL Template Transformation Language (STTL), a generic RDF transformation rule language, independent from the output language. We conceived it as a lightweight syntactic extension to SPARQL and we show how to compile it into standard SPARQL. We present a generic transformation rule engine implementing STTL and several RDF transformers we defined for various output languages, showing STTL’s expressive power.

Related Ontology Subjects/Areas/Topics (6)

Internet Technology		Ontology and the Semantic Web		Web Interfaces and Applications
Metadata and Metamodeling		Web Information Systems and Technologies		XML and Data Management

Feature Matching using CO-Inertia Analysis for People Tracking

Topic: Tracking and Visual Navigation

Authors: Srinidhi Mukanahallipatna Simha , Duc Phu Chau and Francois Bremond

Affiliation: INRIA, France

Digital Object Identifier: 10.5220/0004669502800287

Proceedings of VISAPP 2014/VISIGRAPP, ISBN: 978-989-758-009-3

Quick Abstract

Robust object tracking is a challenging computer vision problem due to dynamic changes in object pose, illumination, appearance and occlusions. Tracking objects between frames requires accurate matching of their features. We investigate real time matching of mobile object features for frame to frame tracking. This paper presents a new feature matching approach between objects for tracking that incorporates one of the multivariate analysis method called Co-Inertia Analysis abbreviated as COIA. This approach is being introduced to compute the similarity between Histogram of Oriented Gradients (HOG) features of the tracked objects. Experiments conducted shows the effectiveness of this approach for mobile object feature tracking.

Related Ontology Subjects/Areas/Topics (3)

Computer Vision, Visualization and Computer Graphics

Motion, Tracking and Stereo Vision

Tracking and Visual Navigation

Group Tracking and Behavior Recognition in Long Video Surveillance Sequences

Topics: Event and Human Activity Recognition; Video Surveillance and Event Detection

Authors: Carolina Gárate , Sofia Zaidenberg , Julien Badie and François Brémond

Affiliation: INRIA, France

Digital Object Identifier: 10.5220/0004682503960402

Proceedings of VISAPP 2014/VISIGRAPP, ISBN: 978-989-758-004-8

Quick Abstract

This paper makes use of recent advances in group tracking and behavior recognition to process large amounts of video surveillance data from an underground railway station and perform a statistical analysis. The most important advantages of our approach are the robustness to process long videos and the capacity to recognize several and different events at once. This analysis automatically brings forward data about the usage of the station and the various behaviors of groups in different hours of the day. This data would be very hard to obtain without an automatic group tracking and behavior recognition method. We present the results and interpretation of one month of processed data from a video surveillance camera in the Torino subway.

Related Ontology Subjects/Areas/Topics (3)

Computer Vision, Visualization and Computer Graphics

Motion, Tracking and Stereo Vision

Video Surveillance and Event Detection

Retrieval System for Person Re-identification

Authors: Sławomir Bąk ¹ , François Brémond ¹ , Vasanth Bathrinarayanan ¹ , Alessandro Capra ² , Davide Giacalone ² , Giuseppe Messina ² and Antonio Buemi ²

Affiliations: ¹ INRIA and STARS group, France , ² STMicroelectronics, Italy

Digital Object Identifier: 10.5220/0004871807320738

Proceedings of PANORAMA 2014/VISIGRAPP, ISBN: 978-989-758-004-8

Quick Abstract

This paper addresses the problem of person re-identification and its application to a real world scenario. We introduce a retrieval system that helps a human operator in browsing a video content. This system is designed for determining whether a given person of interest has already appeared over a network of cameras. In contrast to most of state of the art approaches we do not focus on searching the best strategy for feature matching between camera pairs, but we explore techniques that can perform relatively well in a whole network of cameras. This work is devoted to analyze current state of the art algorithms and technologies, which are currently available on the market. We examine whether current techniques may help a human operator in solving this challenging task. We evaluate our system on the publicly available dataset and demonstrate practical advantages of the proposed solutions.

Towards Reliable Real-time Person Detection

Topics: Machine Learning Technologies for Vision; Features Extraction; Object Detection and Localization

Authors: Silviu-Tudor serban , Srinidhi Mukanahallipatna Simha , Vasanth Bathrinarayanan , Etienne Corvee and Francois Bremond

Affiliation: INRIA Sophia Antipolis - Mediterranee, France

Digital Object Identifier: 10.5220/0004651302320239

Proceedings of VISAPP 2014/VISIGRAPP, ISBN: 978-989-758-004-8

Quick Abstract

We propose a robust real-time person detection system, which aims to serve as solid foundation for developing solutions at an elevated level of reliability. Our belief is that clever handling of input data correlated with efficacious training algorithms are key for obtaining top performance. We introduce a comprehensive training method based on random sampling that compiles optimal classifiers with minimal bias and overfit rate. Building upon recent advances in multi-scale feature computations, our approach attains state-of-the-art accuracy while running at high frame rate.

Related Ontology Subjects/Areas/Topics (3)

Computer Vision, Visualization and Computer Graphics

Features Extraction

Image and Video Analysis

Topic(s):

Title:
ISBN:
Year:
Acronym:
Subject:

Title:
Author:
Affiliation:
Subject:

Name:
Affiliation:
Country:
Conference:
Subject:

Name:
Country:
Subject:

Affiliation

Filter listed results, by