manhattan_sdf icon indicating copy to clipboard operation
manhattan_sdf copied to clipboard

Code for "Neural 3D Scene Reconstruction with the Manhattan-world Assumption" CVPR 2022 Oral

Results 4 manhattan_sdf issues
Sort by recently updated
recently updated
newest added

Hi, thankyou for sharing your wonderful work. I've followed the steps given in the setup. When I proceed further to the training step and run python train_net.py --cfg_file configs/scannet/0050.yaml gpus...

你好: 我在按照文件构建好项目,运行strain_net.py的时候,12G显存一下子就满了,我修改了batch_size也是容易爆满显存,请问你训练的机子都是显存很高的吗? 谢谢!

![屏幕截图 2023-12-30 105953](https://github.com/zju3dv/manhattan_sdf/assets/88564974/1a3ab7a7-1919-4361-b782-5d2079f5eb7d) ![屏幕截图 2023-12-30 110016](https://github.com/zju3dv/manhattan_sdf/assets/88564974/ab8eb903-9ea7-4cce-b35c-4823757a4bb1) Thank you for your work!I‘m confused about the difference in code and paper,why the metric in code higher than paper?

Thank you for your excellent work. I have aligned the extracted mesh with colmap densely matched point cloud, how can I extract the scale and offset values from the colmap...