3dlg-hcvc/M3DRef-CLIP
[ICCV 2023] Multi3DRefer: Grounding Text Description to Multiple 3D Objects
PythonMIT
Issues
- 1
Train on ScanRefer dataset
#19 opened by yihang-xdu - 4
Reproducing Nr3D results in Table 6.
#18 opened by haomengz - 1
Proposal filtering for PointGroup
#17 opened by tony10101105 - 1
How to test on Scanrefer benchmark?
#16 opened by ifrozenwhale - 1
Access the Multi3DRefer test set
#15 opened by haomengz - 1
- 1
Visualization Script
#13 opened by haomengz - 3
Memory cost went too high
#12 opened by Victor-Xiaoyu-Ye - 1
- 2
M3DRef-CLIP on Scanrefer Test Benchmark
#10 opened by ayushjain1144 - 7
- 1
Could not append to config. An item is already at 'ckpt_path'. Either remove + prefix: 'ckpt_path=M3DRef-CLIP_ScanRefer.ckpt' Or add a second + to add or override 'ckpt_path': '++ckpt_path=M3DRef-CLIP_ScanRefer.ckpt'
#8 opened by Andy20178 - 3
- 7
- 2
CLIP text model output. How/Why two outputs word_features & sentence features?
#4 opened by kochsebastian