1 code implementation • 4 Feb 2024 • Matteo Pagliardini, Amirkeivan Mohtashami, Francois Fleuret, Martin Jaggi
The transformer architecture by Vaswani et al. (2017) is now ubiquitous across application domains, from natural language processing to speech processing and image understanding.
1 code implementation • 16 Nov 2023 • Atul Kumar Sinha, Francois Fleuret
We propose an attention-based model to compute an accurate approximation of the EMD that can be used as a training loss for generative models.
1 code implementation • 21 Apr 2023 • Nikolaos Dimitriadis, Francois Fleuret, Pascal Frossard
Continual Learning is an important and challenging problem in machine learning, where models must adapt to a continuous stream of new data without forgetting previously acquired knowledge.
2 code implementations • 14 Jun 2022 • Suraj Srinivas, Kyle Matoba, Himabindu Lakkaraju, Francois Fleuret
To achieve this, we minimize a data-independent upper bound on the curvature of a neural network, which decomposes overall curvature in terms of curvatures and slopes of its constituent layers.
3 code implementations • 7 Mar 2022 • Florian Mai, Arnaud Pannatier, Fabio Fehr, Haolin Chen, Francois Marelli, Francois Fleuret, James Henderson
We find that existing architectures such as MLPMixer, which achieves token mixing through a static MLP applied to each feature independently, are too detached from the inductive biases required for natural language understanding.
no code implementations • CVPR 2021 • Prabhu Teja S, Francois Fleuret
To tackle this problem, we propose a method that reduces the uncertainty of predictions on the target domain data.
no code implementations • ICCV 2021 • Mohammad Mahdi Johari, Camilla Carta, Francois Fleuret
We first propose to use estimated optical flow from ambient information of multiple video frames as a complementary guide for training a single-frame depth estimation network, helping to preserve edges and reduce over-smoothing issues.
1 code implementation • ICLR 2021 • Tatjana Chavdarova, Matteo Pagliardini, Sebastian U. Stich, Francois Fleuret, Martin Jaggi
Generative Adversarial Networks are notoriously challenging to train.
1 code implementation • ICLR 2021 • Suraj Srinivas, Francois Fleuret
This leads us to hypothesize that the highly structured and explanatory nature of input-gradients may be due to the alignment of this class-conditional model $p_{\theta}(x \mid y)$ with that of the ground truth data distribution $p_{\text{data}} (x \mid y)$.
no code implementations • 6 Apr 2020 • Evann Courdier, Francois Fleuret
We propose a change of objective in the segmentation task, and its associated metric that encapsulates this missing information in the following way: We propose to predict the future output segmentation map that will match the future input frame at the time when the network finishes the processing.
no code implementations • 25 Sep 2019 • Prabhu Teja S*, Florian Mai*, Thijs Vogels, Martin Jaggi, Francois Fleuret
There is no consensus yet on the question whether adaptive gradient methods like Adam are easier to use than non-adaptive optimization methods like SGD.
2 code implementations • NeurIPS 2019 • Suraj Srinivas, Francois Fleuret
Our experiments reveal that our method explains model behaviour correctly, and more comprehensively than other methods in the literature.
no code implementations • NeurIPS 2018 • Stepan Tulyakov, Anton Ivanov, Francois Fleuret
End-to-end deep-learning networks recently demonstrated extremely good perfor- mance for stereo matching.
no code implementations • ICML 2018 • Suraj Srinivas, Francois Fleuret
We then rely on this analysis to apply Jacobian matching to transfer learning by establishing equivalence of a recent transfer learning procedure to distillation.
no code implementations • ICCV 2017 • Stepan Tulyakov, Anton Ivanov, Francois Fleuret
Thirdly, it allows to tune deep metric for a particular stereo system, even if ground truth is not available.
no code implementations • ICCV 2017 • Andrii Maksai, Xinchao Wang, Francois Fleuret, Pascal Fua
Many state-of-the-art approaches to multi-object tracking rely on detecting them in each frame independently, grouping detections into short but reliable trajectory segments, and then further grouping them into full trajectories.
1 code implementation • 3 Jul 2017 • Stepan Tulyakov, Anton Ivanov, Nicolas Thomas, Victoria Roloff, Antoine Pommerol, Gabriele Cremonese, Thomas Weigel, Francois Fleuret
There are many geometric calibration methods for "standard" cameras.
no code implementations • ICML 2018 • Cijo Jose, Moustpaha Cisse, Francois Fleuret
It overcomes the ill-conditioning of the recurrent matrix by enforcing soft unitary constraints on the factors.
no code implementations • 3 Dec 2016 • Stepan Tulyakov, Anton Ivanov, Francois Fleuret
The main contribution of our work is a new semi-supervised method for learning deep metrics from unlabeled stereo images, given coarse information about the scenes and the optical system.
1 code implementation • 2 Dec 2016 • Andrii Maksai, Xinchao Wang, Francois Fleuret, Pascal Fua
Many state-of-the-art approaches to people tracking rely on detecting them in each frame independently, grouping detections into short but reliable trajectory segments, and then further grouping them into full trajectories.
no code implementations • CVPR 2016 • Olivier Canevet, Francois Fleuret
We investigate an efficient strategy to collect false positives from very large training sets in the context of object detection.
no code implementations • 1 Mar 2016 • Cijo Jose, Francois Fleuret
We are interested in the large-scale learning of Mahalanobis distances, with a particular focus on person re-identification.
Ranked #110 on Person Re-Identification on Market-1501 (Rank-1 metric)
no code implementations • CVPR 2015 • Timur Bagautdinov, Francois Fleuret, Pascal Fua
We propose a novel approach to computing the probabilities of presence of multiple and potentially occluding objects in a scene from a single depth map.
no code implementations • CVPR 2013 • Raphael Sznitman, Carlos Becker, Francois Fleuret, Pascal Fua
Cascade-style approaches to implementing ensemble classifiers can deliver significant speed-ups at test time.
no code implementations • NeurIPS 2011 • Charles Dubout, Francois Fleuret
Some applications, in particular in computer vision, may involve up to millions of training examples and features.
no code implementations • NeurIPS 2010 • Leonidas Lefakis, Francois Fleuret
The standard strategy for efficient object detection consists of building a cascade composed of several binary classifiers.