no code implementations • 6 Dec 2023 • Doriand Petit, Steve Bourgeois, Dumitru Pavel, Vincent Gay-Bellile, Florian Chabot, Loic Barthe
Recent advances in Neural Fields mostly rely on developing task-specific supervision which often complicates the models.
no code implementations • 5 Nov 2023 • Sandra Kara, Hejer Ammar, Florian Chabot, Quoc-Cuong Pham
This is a critical limitation given the unsupervised setting, where object segments and noise are not distinguishable.
no code implementations • 20 Dec 2022 • Sandra Kara, Hejer Ammar, Florian Chabot, Quoc-Cuong Pham
To address this problem, we propose a fully unsupervised, bottom-up approach, for multiple objects discovery.
no code implementations • 30 May 2022 • Jaonary Rabarisoa, Valentin Belissen, Florian Chabot, Quoc-Cuong Pham
We present a new self-supervised pre-training of Vision Transformers for dense prediction tasks.
no code implementations • 7 Jan 2021 • Abdallah Benzine, Florian Chabot, Bertrand Luvison, Quoc Cong Pham, Cahterine Achrd
It does not need any post-processing to regroup joints since the network predicts a full 3D pose for each bounding box and allows the pose estimation of a possibly large number of people at low resolution.
no code implementations • 4 Nov 2019 • Florian Chabot, Quoc-Cuong Pham, Mohamed Chaouch
In this paper, we propose a new way to efficiently learn a single-shot detector which offers a very good compromise between these two objectives.
no code implementations • CVPR 2017 • Florian Chabot, Mohamed Chaouch, Jaonary Rabarisoa, Céline Teulière, Thierry Chateau
In this paper, we present a novel approach, called Deep MANTA (Deep Many-Tasks), for many-task vehicle analysis from a given image.
Ranked #2 on Vehicle Pose Estimation on KITTI Cars Hard (using extra training data)