Point-GCC: Universal Self-supervised 3D Scene Pre-training via Geometry-Color Contrast

Benchmark Model Rank Results
3d-object-detection-on-sun-rgbd-valPoint-GCC+TR3D+FF#1mAP@0.25: 69.7mAP@0.5: 54.0
3d-object-detection-on-sun-rgbd-valPoint-GCC+TR3D#4mAP@0.25: 67.7mAP@0.5: 51.0
3d-object-detection-on-s3disPoint-GCC+TR3D#3mAP@0.5: 56.7mAP@0.25: 75.1
3d-object-detection-on-scannetv2Point-GCC+TR3D#10mAP@0.5: 59.6mAP@0.25: 73.1