search

zju3dv / NeuralRecon

Code for "NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Video", CVPR 2021 oral
https://zju3dv.github.io/neuralrecon/
Apache License 2.0
2.03k stars 294 forks source link

Single-layered mesh generation. #46

Open xincoder opened 3 years ago

xincoder commented 3 years ago

Thank you for the great work and for sharing the code. How should we modify the code to generate "single-layered mesh" as shown in Figure 2 in Supplementary Material? Thank you very much.

boother commented 2 years ago

@xincoder Did you find solution ? :)

HaFred commented 1 year ago

The reason why the output mesh is in a double-layered fashion is that the gt tsdf_vol is generated as double-layered in this work. To convert it into single-layered, I guess either to change the gt tsdf_vol and re-train it, or using extracted depth to conduct TSDF fusion and dump the mesh.