Full Surround Monodepth From Multiple Cameras
| Autor: | Vitor Guizilini, Igor Vasiljevic, Rares Ambrus, Greg Shakhnarovich, Adrien Gaidon |
|---|---|
| Rok vydání: | 2022 |
| Předmět: |
FOS: Computer and information sciences
Human-Computer Interaction Control and Optimization Artificial Intelligence Control and Systems Engineering Computer Vision and Pattern Recognition (cs.CV) Mechanical Engineering Computer Science - Computer Vision and Pattern Recognition ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Biomedical Engineering Computer Vision and Pattern Recognition Computer Science Applications |
| Zdroj: | IEEE Robotics and Automation Letters. 7:5397-5404 |
| ISSN: | 2377-3774 |
| DOI: | 10.1109/lra.2022.3150884 |
| Popis: | Self-supervised monocular depth and ego-motion estimation is a promising approach to replace or supplement expensive depth sensors such as LiDAR for robotics applications like autonomous driving. However, most research in this area focuses on a single monocular camera or stereo pairs that cover only a fraction of the scene around the vehicle. In this work, we extend monocular self-supervised depth and ego-motion estimation to large-baseline multi-camera rigs. Using generalized spatio-temporal contexts, pose consistency constraints, and carefully designed photometric loss masking, we learn a single network generating dense, consistent, and scale-aware point clouds that cover the same full surround 360 degree field of view as a typical LiDAR scanner. We also propose a new scale-consistent evaluation metric more suitable to multi-camera settings. Experiments on two challenging benchmarks illustrate the benefits of our approach over strong baselines. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|