Zobrazeno 1 - 10
of 108
pro vyhledávání: '"P, Filliat"'
The recent development of foundation models for monocular depth estimation such as Depth Anything paved the way to zero-shot monocular depth estimation. Since it returns an affine-invariant disparity map, the favored technique to recover the metric d
Externí odkaz:
http://arxiv.org/abs/2412.14103
Accurate trajectory forecasting is crucial for the performance of various systems, such as advanced driver-assistance systems and self-driving vehicles. These forecasts allow to anticipate events leading to collisions and, therefore, to mitigate them
Externí odkaz:
http://arxiv.org/abs/2403.17678
Temporal Difference (TD) algorithms are widely used in Deep Reinforcement Learning (RL). Their performance is heavily influenced by the size of the neural network. While in supervised learning, the regime of over-parameterization and its benefits are
Externí odkaz:
http://arxiv.org/abs/2310.05518
Autor:
Franchi, Gianni, Hariat, Marwane, Yu, Xuanlong, Belkhir, Nacim, Manzanera, Antoine, Filliat, David
Current deep neural networks (DNNs) for autonomous driving computer vision are typically trained on specific datasets that only involve a single type of data and urban scenes. Consequently, these models struggle to handle new objects, noise, nighttim
Externí odkaz:
http://arxiv.org/abs/2309.15751
Publikováno v:
Proceedings of The 2nd Conference on Lifelong Learning Agents, PMLR 232 (2023) 658-682
End-to-end reinforcement learning on images showed significant progress in the recent years. Data-based approach leverage data augmentation and domain randomization while representation learning methods use auxiliary losses to learn task-relevant fea
Externí odkaz:
http://arxiv.org/abs/2306.08537
Autor:
Franchi, Gianni, Yu, Xuanlong, Bursuc, Andrei, Aldea, Emanuel, Dubuisson, Severine, Filliat, David
Predictive uncertainty estimation is essential for deploying Deep Neural Networks in real-world autonomous systems. However, most successful approaches are computationally intensive. In this work, we attempt to address these challenges in the context
Externí odkaz:
http://arxiv.org/abs/2207.10130
Autor:
Franchi, Gianni, Yu, Xuanlong, Bursuc, Andrei, Tena, Angel, Kazmierczak, Rémi, Dubuisson, Séverine, Aldea, Emanuel, Filliat, David
Predictive uncertainty estimation is essential for safe deployment of Deep Neural Networks in real-world autonomous systems. However, disentangling the different types and sources of uncertainty is non trivial for most datasets, especially since ther
Externí odkaz:
http://arxiv.org/abs/2203.01437
Several metrics exist to quantify the similarity between images, but they are inefficient when it comes to measure the similarity of highly distorted images. In this work, we propose to empirically investigate perceptual metrics based on deep neural
Externí odkaz:
http://arxiv.org/abs/2202.08692
Autor:
Chen, Zhaorun, Fan, Siqi, Tan, Yuan, Gong, Liang, Chen, Binhao, Sun, Te, Filliat, David, Díaz-Rodríguez, Natalia, Liu, Chengliang
While the rapid progress of deep learning fuels end-to-end reinforcement learning (RL), direct application, especially in high-dimensional space like robotic scenarios still suffers from low sample efficiency. Therefore State Representation Learning
Externí odkaz:
http://arxiv.org/abs/2109.08642
Affordances are the possibilities of actions the environment offers to the individual. Ordinary objects (hammer, knife) usually have many affordances (grasping, pounding, cutting), and detecting these allow artificial agents to understand what are th
Externí odkaz:
http://arxiv.org/abs/2107.02095