Sparse LiDAR and Stereo Fusion (SLS-Fusion) for Depth Estimation and 3D Object Detection
| Autor: | Pierre Duthon, Sergio A. Velastin, Louahdi Khoudour, Nguyen Anh Minh Mai, Alain Crouzil |
|---|---|
| Přispěvatelé: | CROUZIL, Alain, Institution of Engineering and Technology (IET), CoMputational imagINg anD viSion (IRIT-MINDS), Institut de recherche en informatique de Toulouse (IRIT), Université Toulouse 1 Capitole (UT1), Université Fédérale Toulouse Midi-Pyrénées-Université Fédérale Toulouse Midi-Pyrénées-Université Toulouse - Jean Jaurès (UT2J)-Université Toulouse III - Paul Sabatier (UT3), Université Fédérale Toulouse Midi-Pyrénées-Centre National de la Recherche Scientifique (CNRS)-Institut National Polytechnique (Toulouse) (Toulouse INP), Université Fédérale Toulouse Midi-Pyrénées-Université Toulouse 1 Capitole (UT1), Université Fédérale Toulouse Midi-Pyrénées, Centre d'Etudes et d'Expertise sur les Risques, l'Environnement, la Mobilité et l'Aménagement - Equipe-projet STI (Cerema Equipe-projet STI), Centre d'Etudes et d'Expertise sur les Risques, l'Environnement, la Mobilité et l'Aménagement (Cerema), School of Electronic Engineering and Computer Science (EECS), Queen Mary University of London (QMUL), Carlos III University of Madrid |
| Rok vydání: | 2021 |
| Předmět: |
Autonomous vehicle
Computer science LiDAR stereo fusion ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Depth completion [INFO.INFO-CV] Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] 11. Sustainability Computer vision Informática Artificial neural network business.industry Pseudo lidar [INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV] 2D to 3D conversion Pseudo LiDAR Object detection 3D object detection Lidar Fuse (electrical) RGB color model Lidar stereo fusion Artificial intelligence business Focus (optics) Stereo camera |
| Zdroj: | 11th International Conference on Pattern Recognition Systems (ICPRS 2021) e-Archivo: Repositorio Institucional de la Universidad Carlos III de Madrid Universidad Carlos III de Madrid (UC3M) e-Archivo. Repositorio Institucional de la Universidad Carlos III de Madrid instname |
| DOI: | 10.1049/icp.2021.1442 |
| Popis: | Procedings in: 11th International Conference on Pattern Recognition Systems (ICPRS-21), conference paper, 17-19 mar, 2021, Universidad de Talca, Curicó, Chile. The ability to accurately detect and localize objects is recognized as being the most important for the perception of self-driving cars. From 2D to 3D object detection, the most difficult is to determine the distance from the ego-vehicle to objects. Expensive technology like LiDAR can provide a precise and accurate depth information, so most studies have tended to focus on this sensor showing a performance gap between LiDAR-based methods and camera-based methods. Although many authors have investigated how to fuse LiDAR with RGB cameras, as far as we know there are no studies to fuse LiDAR and stereo in a deep neural network for the 3D object detection task. This paper presents SLS-Fusion, a new approach to fuse data from 4-beam LiDAR and a stereo camera via a neural network for depth estimation to achieve better dense depth maps and thereby improves 3D object detection performance. Since 4-beam LiDAR is cheaper than the well-known 64-beam LiDAR, this approach is also classified as a low-cost sensors-based method. Through evaluation on the KITTI benchmark, it is shown that the proposed method significantly improves depth estimation performance compared to a baseline method. Also when applying it to 3D object detection, a new state of the art on low-cost sensor based method is achieved. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|