Rui Li

thank you a lot！ TWJianNuo 于2020年11月11日周三 下午9:58写道： > According to their paper > , > it is pretrained on Cityscape and finetuned with kitti 200 semantics > segmentation label. >...

Thank you for your reply~ I double checked the page and found the kitti model on another [branch](https://github.com/NVIDIA/semantic-segmentation/tree/sdcnet) of the project. Thank you for the other link of the extra...

Thank you a lot! I have one more question, then how to locate the corresponding multiple depth-semantic pairs when given a point at $x$? TWJianNuo 于2020年7月28日周二 下午8:46写道： > Hi! >...

That's a very detailed explanation, thank you a lot! TWJianNuo 于2020年7月28日周二 下午10:24写道： > For one semantic edge pixel, it will be paired with a nearest depth edge > pixel. If...

[update] Even with the same environment as suggested in the repository, I also get similar evaluation results using the provided model

> Hi, thanks for your wonderful work! I'm trying to repeat the quantitative results in your paper on a single A100 GPU. I am able to generate the tsdf fusion...

Hi, Thanks for your attention. We use `plyfile` lib to generate the 3D voxels. The code was released and you can check it out here. https://github.com/ruili3/Know-Your-Neighbors/blob/733f18597c86d9fdeafe2249509650533e4f6ead/scripts/gen_kitti360_voxel.py#L269 Feel free if you...

Hi, you can follow the demo in DUSt3R to compute the depth map: https://github.com/naver/dust3r/blob/69192aa322d279438390c109b8b85d5b859b5cdd/dust3r/demo.py#L168 Note that DUSt3R's depth map is up to a scale to the real-world depth, so you...

Hey, do you mean using rendered images from DUSt3R/MASt3R for image-to-3D generation? Currently no, but the recent InstatSplat or Splatt3R may be used for this purpose. Best

Thanks, 1) I would like the multiple depth maps from an opaque object, and the renderer performs rasterization through all ray-surface intersections. I tried PyTorch3D for this purpose in [LaRI](https://ruili3.github.io/lari/index.html);...