Using Cartesian voxelization and a positive-pair mining module to exploit unsynced LiDAR-image data improves image-to-LiDAR representation learning, setting new state-of-the-art on nuScenes segmentation and KITTI detection.
In: Proceedings of the IEEE/CVF International Conference on Computer Vision
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
The Devil is in the Details: Simple Remedies for Image-to-LiDAR Representation Learning
Using Cartesian voxelization and a positive-pair mining module to exploit unsynced LiDAR-image data improves image-to-LiDAR representation learning, setting new state-of-the-art on nuScenes segmentation and KITTI detection.