AndyYuan96

dynamic voxelization generate negative index

1

I use dynamic voxelization to train nus model, I find sometime's voxelize will generate negative index, which make index operation out of range, even if I already use PointsRangeFIlter. In...

problem about waymo dataset lidar saving

2

for waymo dataset's lidar saving, why not saving the image position which is cp_points in code for each lidar point, as waymo not only use calibration to convert lidar point...

Hi, thanks you for so great work, I just wonder where you will give the nuscene result in the future, as your paper's result is so low in other two...

some question about position encoding in depth axis

1

Hi, thanks for sharing so wonderful work, after reading the paper, I have some maybe stupid question, how about set D=1? With my understanding, for each position in different view...

hi，nazhao，I‘m trying semi-supervised learning method on 3d object detection on outdoor dataset，I just wonder How did you think of using 100% trainset both as labeled data and unlabeled data，as in...

some question about VSA encoder code

2

Hi, thanks for sharing the worke. In VSA encoder, code is below: x = self.pre_mlp(inp) # encoder attn = torch_scatter.scatter_softmax(self.score(x), inverse, dim=0) dot = (attn[:, :, None] * x.view(-1, 1,...

[Feature] GT Augmentation Fade Strategy

1

Thanks for your contribution and we appreciate it a lot. The following instructions would make your pull request more healthy and more easily get feedback. If you do not understand...

inference speed in paper

1

Hi, you say the transfusion speed is 156.6 ms / frame in your paper, but in transfusion paper, the speed is 265.9ms/frame. So your guys just change gpu from v100...

Hi, tianwei, I just wonder does ensembling of different voxelsize models and emsembling different rotation output use some function, which is nms.

how is the performance?

1

Hi, ed-fish, thanks for sharing your implementation of the paper, I just wonder how is the performance of your implementation, does it is comparable with original paper?