chaytonmin / deepmvs Goto Github PK

您好，我已经按照之前那个docker的命令安装好了环境，但是我想知道如何进入这个docker的环境以及视频中的start_docker_mvs.sh，mvs_cascade.pl和colmap.sh文件在哪里，我在github开源的那个代码中没有找到，谢谢！

Hello, I have installed the environment according to the previous docker command, but I want to know how to enter the docker environment and where the start_docker_mvs.sh, mvs_cascade.pl and colmap.sh files in the video are. I can't find them in the open source code of GitHub. Thank you！

I replaced colmap with hloc as your second "to do" in readme, and got a little optimization.
But when I use casmvsnet and depth2dmap to generate *.dmap instead of openmvs, I got a chaotic result. The target is a person but a cube is generated. I am wondering how your model is trained and how can I get correct dmap.

Hello, is there any difference between cascade_pl_4.12_colmap/eval.py and cascade_pl/eval.py？

how to install colmap?

I've tested the code and managed to generate multiple pfm files. However, after running openmvs, I realized that these generated pfm files were not adopted at all. The final results still depend on the estimated dmp files by openmvs's DensifyPointCloud.exe, so the generated result have nothing to do with the pfm files?

Please confirm this. Thanks!

How can I know which picture is used as the reference image for the reconstructed point cloud?

Is the content in Docker not updated?

Hi, thank you for contributing to this repository. It's worth using deep learning approaches to improve the quality of the sparse point cloud before densification and the rest. How did you manage the output from CasMVSNet or another depth inferring method in order to use it in the OpenMVS step? Any data modification or libraries required? Thank you.