no code implementations • 4 Nov 2023 • Abid Ali, Ashish Marisetty, Francois Bremond
To address these challenges, we propose AgeFormer which utilizes spatio-temporal information on the dynamics of the entire body dominating face-based methods for age classification.
no code implementations • 12 Sep 2023 • Mohammed Guermal, Francois Bremond, Rui Dai, Abid Ali
By combining action anticipation and online action detection, our approach can cover the missing dependencies of future information in online action detection.
no code implementations • 28 Aug 2023 • Di Yang, Yaohui Wang, Antitza Dantcheva, Quan Kong, Lorenzo Garattoni, Gianpiero Francesca, Francois Bremond
In this context, we propose Latent Action Composition (LAC), a novel self-supervised framework aiming at learning from synthesized composable motions for skeleton-based action segmentation.
no code implementations • 10 May 2023 • Di Yang, Yaohui Wang, Quan Kong, Antitza Dantcheva, Lorenzo Garattoni, Gianpiero Francesca, Francois Bremond
Self-supervised video representation learning aimed at maximizing similarity between different temporal segments of one video, in order to enforce feature persistence over time.
no code implementations • 19 Jan 2023 • Snehashis Majhi, Rui Dai, Quan Kong, Lorenzo Garattoni, Gianpiero Francesca, Francois Bremond
Video anomaly detection in surveillance systems with only video-level labels (i. e. weakly-supervised) is challenging.
1 code implementation • 2 Jan 2023 • Hao Chen, Yaohui Wang, Benoit Lagadec, Antitza Dantcheva, Francois Bremond
This work focuses on unsupervised representation learning in person re-identification (ReID).
no code implementations • ICCV 2023 • Di Yang, Yaohui Wang, Antitza Dantcheva, Quan Kong, Lorenzo Garattoni, Gianpiero Francesca, Francois Bremond
In this context, we propose Latent Action Composition (LAC), a novel self-supervised framework aiming at learning from synthesized composable motions for skeleton-based action segmentation.
1 code implementation • 31 Aug 2022 • Di Yang, Yaohui Wang, Antitza Dantcheva, Lorenzo Garattoni, Gianpiero Francesca, Francois Bremond
Current self-supervised approaches for skeleton action representation learning often focus on constrained scenarios, where videos and skeleton data are recorded in laboratory settings.
no code implementations • 19 Aug 2022 • Indu Joshi, Marcel Grimmer, Christian Rathgeb, Christoph Busch, Francois Bremond, Antitza Dantcheva
This survey is intended for researchers and practitioners in the field of human analysis.
no code implementations • 20 Apr 2022 • Mohammed Guermal, Rui Dai, Francois Bremond
In this work, we present an end-to-end network: THORN, that can leverage important human-object and object-object interactions to predict actions.
no code implementations • 17 Mar 2022 • Yaohui Wang, Di Yang, Francois Bremond, Antitza Dantcheva
Specifically, motion in generated video is constructed by linear displacement of codes in the latent space.
1 code implementation • 12 Mar 2022 • Hao Chen, Benoit Lagadec, Francois Bremond
Existing unsupervised person re-identification (ReID) methods focus on adapting a model trained on a source domain to a fixed target domain.
Unsupervised Domain Adaptation Unsupervised Person Re-Identification
no code implementations • 22 Dec 2021 • Tanay Agrawal, Dhruv Agarwal, Michal Balazia, Neelabh Sinha, Francois Bremond
Personality computing and affective computing have gained recent interest in many research areas.
1 code implementation • CVPR 2022 • Rui Dai, Srijan Das, Kumara Kahatapitiya, Michael S. Ryoo, Francois Bremond
Action detection is an essential and challenging task, especially for densely labelled datasets of untrimmed videos.
Ranked #2 on Action Detection on TSU
no code implementations • 26 Oct 2021 • Rui Dai, Srijan Das, Francois Bremond
Action detection is an essential and challenging task, especially for densely labelled datasets of untrimmed videos.
Ranked #2 on Action Detection on Multi-THUMOS
1 code implementation • 10 Oct 2021 • Neelabh Sinha, Michal Balazia, Francois Bremond
3D gaze estimation is about predicting the line of sight of a person in 3D space.
1 code implementation • ICLR 2022 • Yaohui Wang, Di Yang, Francois Bremond, Antitza Dantcheva
Deviating from such models, we here introduce Latent Image Animator (LIA), a self-supervised auto-encoder that evades need for structure representation.
no code implementations • 20 Aug 2021 • Snehashis Majhi, Srijan Das, Francois Bremond, Ratnakar Dash, Pankaj Kumar Sa
Thinking of a fully automatized surveillance system, which is capable of both detecting and classifying the anomalies that need immediate actions, a joint anomaly detection and classification method is a pressing need.
no code implementations • ICCV 2021 • Rui Dai, Srijan Das, Francois Bremond
On the other hand, sequence-level distillation encourages the student to learn the temporal knowledge from the teacher, which consists of transferring the Global Contextual Relations and the Action Boundary Saliency.
1 code implementation • 19 Jul 2021 • Di Yang, Yaohui Wang, Antitza Dantcheva, Lorenzo Garattoni, Gianpiero Francesca, Francois Bremond
This is achieved by learning an optimal dependency matrix from the uniform distribution based on a multi-head attention mechanism.
Ranked #1 on Skeleton Based Action Recognition on UPenn Action
1 code implementation • 17 May 2021 • Srijan Das, Rui Dai, Di Yang, Francois Bremond
But the cost of computing 3D poses from RGB stream is high in the absence of appropriate sensors.
Ranked #10 on Action Recognition on NTU RGB+D 120 (using extra training data)
1 code implementation • ICCV 2021 • Hao Chen, Benoit Lagadec, Francois Bremond
Then, we use similarity scores as soft pseudo labels to enhance the consistency between augmented and original views, which makes our model more robust to augmentation perturbations.
no code implementations • 9 Feb 2021 • Michal Balazia, S L Happy, Francois Bremond, Antitza Dantcheva
Face recognition has been widely accepted as a means of identification in applications ranging from border control to security in the banking sector.
no code implementations • 8 Jan 2021 • Yaohui Wang, Francois Bremond, Antitza Dantcheva
We design the architecture of InMoDeGAN-generator in accordance to proposed Linear Motion Decomposition, which carries the assumption that motion can be represented by a dictionary, with related vectors forming an orthogonal basis in the latent space.
1 code implementation • 5 Jan 2021 • Rui Dai, Srijan Das, Luca Minciullo, Lorenzo Garattoni, Gianpiero Francesca, Francois Bremond
Previous action detection methods fail in selecting the key temporal information in long videos.
Ranked #1 on Action Detection on TSU
2 code implementations • CVPR 2021 • Hao Chen, Yaohui Wang, Benoit Lagadec, Antitza Dantcheva, Francois Bremond
In this context, we propose a mesh-based view generator.
1 code implementation • 27 Nov 2020 • Hao Chen, Benoit Lagadec, Francois Bremond
The objective of unsupervised person re-identification (Re-ID) is to learn discriminative features without labor-intensive identity annotations.
Unsupervised Domain Adaptation Unsupervised Person Re-Identification
1 code implementation • 10 Nov 2020 • Di Yang, Rui Dai, Yaohui Wang, Rupayan Mallick, Luca Minciullo, Gianpiero Francesca, Francois Bremond
Taking advantage of human pose data for understanding human activities has attracted much attention these days.
1 code implementation • 28 Oct 2020 • Rui Dai, Srijan Das, Saurav Sharma, Luca Minciullo, Lorenzo Garattoni, Francois Bremond, Gianpiero Francesca
Therefore, we propose a new baseline method for activity detection to tackle the novel challenges provided by our dataset.
1 code implementation • ECCV 2020 • Srijan Das, Saurav Sharma, Rui Dai, Francois Bremond, Monique Thonnat
The 2 key components of this VPN are a spatial embedding and an attention network.
Ranked #6 on Action Classification on Toyota Smarthome dataset (using extra training data)
1 code implementation • CVPR 2020 • Yaohui Wang, Piotr Bilinski, Francois Bremond, Antitza Dantcheva
Creating realistic human videos entails the challenge of being able to simultaneously generate both appearance, as well as motion.
no code implementations • 1 Feb 2018 • Srijan Das, Michal Koperski, Francois Bremond, Gianpiero Francesca
In this paper, we propose to improve the traditional use of RNNs by employing a many to many model for video classification.
no code implementations • 20 Jul 2016 • Furqan M. Khan, Francois Bremond
Current state-of-the-art methods often address the problem by relying on supervised learning of similarity metrics or ranking functions to implicitly model appearance transformation between cameras for each camera pair, or group, in the system.