PointTransformerV2
PointTransformerV2 copied to clipboard
Pointcept
[NeurIPS'22] An official PyTorch implementation of PTv2.
hello, I think your work is really excellent. Well, it is a little difficult for me to train s3dis, could you offer me your checkpoint? Thank you so much.
When I ran the ptv2 code before, it was fine and could be trained and tested as normal. However, a few days later, whenever "import pcr.models", the above error always...
Hi,here is a problem! multiprocessing.pool.RemoteTraceback: """ Traceback (most recent call last): File "/home/jsjlab/anaconda3/envs/pcr/lib/python3.8/multiprocessing/pool.py", line 125, in worker result = (True, func(*args, **kwds)) File "/home/jsjlab/anaconda3/envs/pcr/lib/python3.8/multiprocessing/pool.py", line 51, in starmapstar return list(itertools.starmap(args[0],...
GVA is proposed in PTV2 as a new method, but the implementation is equal as Vector Attention with shared plane in PTV1. below are the comparisons: PTV1: w = self.softmax(w)...
multiprocessing.pool.RemoteTraceback: """ Traceback (most recent call last): File "/home/xqq/developer/ide/anaconda3/envs/pcr/lib/python3.8/multiprocessing/pool.py", line 125, in worker result = (True, func(*args, **kwds)) File "/home/xqq/developer/ide/anaconda3/envs/pcr/lib/python3.8/multiprocessing/pool.py", line 51, in starmapstar return list(itertools.starmap(args[0], args[1])) File "/home/xqq/developer/code/python/PointTransformerV2-main/pcr/datasets/preprocessing/s3dis/preprocess_s3dis.py", line...
Hi I wanted to clarify the below comment in your repo. The original S3DIS data contains some bugs data need manually fix it. xxx^@xxx when i try to run the...
hello, I think your work is really excellent. Well, it is a little difficult for me to scannet visualization , could you offer me your code? Thank you so much.
Thanks for sharing. I had some problems when I submitted my testing results to ScanNet. I had a competitive performance on the validation set, such as 74.0% mIoU, but only...
First of all, thank you so much for this great work! Having multiple models all in one place saves a lot of time for others when benchmarking! I'm wondering about...
Hi, Thanks to your contributes. I want to train PTv1, dataset=shapenet, for part segmentation. Could you give me a running script for this please? Thank you :)
