no code implementations • 18 Mar 2024 • Quentin Herau, Moussab Bennehar, Arthur Moreau, Nathan Piasco, Luis Roldao, Dzmitry Tsishkou, Cyrille Migniot, Pascal Vasseur, Cédric Demonceaux
We introduce 3DGS-Calib, a new calibration method that relies on the speed and rendering accuracy of 3D Gaussian Splatting to achieve multimodal spatiotemporal calibration that is accurate, robust, and with a substantial speed-up compared to methods relying on implicit neural representations.
no code implementations • 15 Mar 2024 • Hiba Dahmani, Moussab Bennehar, Nathan Piasco, Luis Roldao, Dzmitry Tsishkou
To tackle this, we extend over 3D Gaussian Splatting to handle unstructured image collections.
no code implementations • 15 Mar 2024 • Hala Djeghim, Nathan Piasco, Moussab Bennehar, Luis Roldão, Dzmitry Tsishkou, Désiré Sidibé
SCILLA's hybrid architecture models two separate implicit fields: one for the volumetric density and another for the signed distance to the surface.
no code implementations • 14 Mar 2024 • Thang-Anh-Quan Nguyen, Luis Roldão, Nathan Piasco, Moussab Bennehar, Dzmitry Tsishkou
The task of separating dynamic objects from static environments using NeRFs has been widely studied in recent years.
no code implementations • 27 Nov 2023 • Quentin Herau, Nathan Piasco, Moussab Bennehar, Luis Roldão, Dzmitry Tsishkou, Cyrille Migniot, Pascal Vasseur, Cédric Demonceaux
In this paper, we leverage the ability of Neural Radiance Fields (NeRF) to represent different sensors modalities in a common volumetric representation to achieve robust and accurate spatio-temporal sensor calibration.
no code implementations • 26 May 2023 • Fusang Wang, Arnaud Louys, Nathan Piasco, Moussab Bennehar, Luis Roldão, Dzmitry Tsishkou
Neural Radiance Fields (NeRF) enable 3D scene reconstruction from 2D images and camera poses for Novel View Synthesis (NVS).
no code implementations • ICCV 2023 • Arthur Moreau, Nathan Piasco, Moussab Bennehar, Dzmitry Tsishkou, Bogdan Stanciulescu, Arnaud de La Fortelle
Beyond novel view synthesis, Neural Radiance Fields are useful for applications that interact with the real world.
no code implementations • 6 Mar 2023 • Quentin Herau, Nathan Piasco, Moussab Bennehar, Luis Roldão, Dzmitry Tsishkou, Cyrille Migniot, Pascal Vasseur, Cédric Demonceaux
With the recent advances in autonomous driving and the decreasing cost of LiDARs, the use of multimodal sensor systems is on the rise.
no code implementations • 10 Oct 2022 • Caio Azevedo, Thomas Gilles, Stefano Sabatini, Dzmitry Tsishkou
Inspired by recent developments regarding the application of self-supervised learning (SSL), we devise an auxiliary task for trajectory prediction that takes advantage of map-only information such as graph connectivity with the intent of improving map comprehension and generalization.
no code implementations • 9 Aug 2022 • Yann Koeberle, Stefano Sabatini, Dzmitry Tsishkou, Christophe Sabourin
In this work, we show that a trade-off exists between imitating human driving and maintaining safety when learning driving policies.
no code implementations • 15 May 2022 • Thomas Gilles, Stefano Sabatini, Dzmitry Tsishkou, Bogdan Stanciulescu, Fabien Moutarde
While a lot of work has been carried on developing trajectory prediction methods, and various datasets have been proposed for benchmarking this task, little study has been done so far on the generalizability and the transferability of these methods across dataset.
no code implementations • 5 May 2022 • Arthur Moreau, Thomas Gilles, Nathan Piasco, Dzmitry Tsishkou, Bogdan Stanciulescu, Arnaud de La Fortelle
We propose a novel learning-based formulation for visual localization of vehicles that can operate in real-time in city-scale environments.
no code implementations • 13 Oct 2021 • Arthur Moreau, Nathan Piasco, Dzmitry Tsishkou, Bogdan Stanciulescu, Arnaud de La Fortelle
Neural Radiance Fields (NeRF) have recently demonstrated photo-realistic results for the task of novel view synthesis.
no code implementations • ICLR 2022 • Thomas Gilles, Stefano Sabatini, Dzmitry Tsishkou, Bogdan Stanciulescu, Fabien Moutarde
In this paper, we propose THOMAS, a joint multi-agent trajectory prediction framework allowing for an efficient and consistent prediction of multi-agent multi-modal trajectories.
Ranked #6 on Trajectory Prediction on nuScenes
no code implementations • 4 Sep 2021 • Thomas Gilles, Stefano Sabatini, Dzmitry Tsishkou, Bogdan Stanciulescu, Fabien Moutarde
In this paper, we propose GOHOME, a method leveraging graph representations of the High Definition Map and sparse projections to generate a heatmap output representing the future position probability distribution for a given agent in a traffic scene.
Ranked #1 on Trajectory Prediction on INTERACTION Dataset - Validation (minFDE6 metric)
1 code implementation • 23 May 2021 • Thomas Gilles, Stefano Sabatini, Dzmitry Tsishkou, Bogdan Stanciulescu, Fabien Moutarde
In this paper, we propose HOME, a framework tackling the motion forecasting problem with an image output representing the probability distribution of the agent's future location.
Ranked #32 on Motion Forecasting on Argoverse CVPR 2020
no code implementations • 19 Mar 2021 • Arthur Moreau, Nathan Piasco, Dzmitry Tsishkou, Bogdan Stanciulescu, Arnaud de La Fortelle
In this setup, structure-based methods require a large database, and we show that our proposal is a reliable alternative, achieving 29cm median error in a 1. 9km loop in a busy urban area
Ranked #2 on Camera Localization on Oxford RobotCar Full