Make Your ViT-based Multi-view 3D Detectors Faster via Token Compression
Open paper
Benchmark
Model
Rank
Results
3d-object-detection-on-nuscenes
StreamPETR + ToC3D-Faster (ViT-L) (extra data)
#35
NDS: 0.626
mAP: 0.542
Rank counts only results with a code link.