slothfulxtx

Maybe the `batch_size` is 16\*4 = 64. I run the code with `batch_size=4*4` , and the accuracy on the CUB_200_2011 dataset is only 90.9\%. After changing the `batch_size` to 4\*8...

I have trained almost all model on completion3d dataset, using the scripts provided in this repo. However, I can't reproduce the perfomance of most of them on the leaderboard. Thus,...

Hello, everyone, we implement a diff-gauss-rasterization repo which supports more than alpha, depth, normal maps rendering. You can find our code here https://github.com/slothfulxtx/diff-gaussian-rasterization

Here's my implementation, hope it can be helpful! [https://github.com/slothfulxtx/diff-gaussian-rasterization](https://github.com/slothfulxtx/diff-gaussian-rasterization)

We also support both alpha map and depth map now! (*^_^*)

Hi, I've uploaded all my code and checkpoints on github, thus it should be easy to reproduce our results. I've just tested my code and checkpoint on our server ```...

Actually, MBPTrack reuses previous prediction results when the tracked target is missing. If the prediction of MBPTrack is not accurate enough, it may lose targets for the following frames. (You...

Hi, sorry to hear that. Based on my memory, I use only 2 3090Ti GPUs for training, under the conda environment provided by the `freeze.yml`. No more tricks were used...

Hi, We use pointnet2 repo for its point-based operations. You can find the dependencies [here](https://github.com/erikwijmans/Pointnet2_PyTorch)

Can you visualize the tracking results? We provide a simple visualization script [here](https://github.com/slothfulxtx/cxtrack3d/blob/main/visualize.py). Maybe you need make some modification for NuScenes dataset. As far as i know, compared with kitti...