Roses are Red, Violets are Blue… But Should VQA expect Them To?

Autor: Grigory Antipov, Christian Wolf, Corentin Kervadec, Moez Baccouche

Publikováno v: CVPR
HAL

Visual Question Answering (VQA) models are notorious for their tendency to rely on dataset biases.The large and unbalanced diversity of questions and concepts involved in VQA and the lack of high standard annotated data tend to prevent models from le

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::67fd96af15e227493c4b56fa70febeed
https://doi.org/10.1109/cvpr46437.2021.00280

MFAS: Multimodal Fusion Architecture Search

Autor: Moez Baccouche, Juan-Manuel Perez-Rua, Valentin Vielzeuf, Frédéric Jurie, Stéphane Pateux

Publikováno v: CVPR
CVPR 2019
CVPR 2019, Jun 2019, Long Beach, United States

We tackle the problem of finding good architectures for multimodal classification problems. We propose a novel and generic search space that spans a large number of possible fusion architectures. In order to find an optimal architecture for a given d

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8c807147acda165f9b075c6384401ee6
https://doi.org/10.1109/cvpr.2019.00713

Boosting Cross-Age Face Verification via Generative Age Normalization

Autor: Grigory Antipov, Jean-Luc Dugelay, Moez Baccouche

Publikováno v: International Joint Conference on Biometrics
International Joint Conference on Biometrics, Oct 2017, Denver, United States
IJCB

International audience; Despite the tremendous progress in face verification performance as a result of Deep Learning, the sensitivity to human age variations remains an Achilles' heel of the majority of the contemporary face verification software. A

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b5eb700e64a3e12ed21cbf247fe47270
https://hal.archives-ouvertes.fr/hal-01617381

Evaluation of video activity localizations integrating quality and quantity measurements

Autor: Charles-Edmond Bichot, Julien Mille, Oya Celiktutan, Bulent Sankur, Eric Lombardi, Emmanuel Dellandréa, Christophe Garcia, Gonen Eren, Christian Wolf, Moez Baccouche, Mingyuan Jiu, Emre Dogan

Publikováno v: Computer Vision and Image Understanding
Computer Vision and Image Understanding, Elsevier, 2014, 127, pp.14-30. ⟨10.1016/j.cviu.2014.06.014⟩

International audience; Evaluating the performance of computer vision algorithms is classically done by reporting classification error or accuracy, if the problem at hand is the classification of an object in an image, the recognition of an activity

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::136374b38e616029513528029beca0ce
https://doi.org/10.1016/j.cviu.2014.06.014

Blind Super-Resolution with Deep Convolutional Neural Networks

Autor: Moez Baccouche, Christophe Garcia, Clément Peyrard

Publikováno v: Artificial Neural Networks and Machine Learning – ICANN 2016 ISBN: 9783319447803
ICANN (2)
Artificial Neural Networks and Machine Learning
ICANN 2016: 25th International Conference on Artificial Neural Networks
ICANN 2016: 25th International Conference on Artificial Neural Networks, Sep 2016, Barcelona, Spain. pp.161--169, ⟨10.1007/978-3-319-44781-0_20⟩

International audience; Example-based methods have demonstrated their ability to perform well for Single Image Super-Resolution (SR). While very efficient when a single image formation model (non-blind) is assumed for the low-resolution (LR) observat

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8636de4371d11c17fe631a817c62b0ce
https://doi.org/10.1007/978-3-319-44781-0_20