yanghb22-fdu / Hi3D-Official

[MM24] Official codes and datasets for ACM MM24 paper "Hi3D: Pursuing High-Resolution Image-to-3D Generation with Video Diffusion Models".
MIT License
215 stars 12 forks source link

3D Mesh Extraction #15

Open wondering516 opened 1 month ago

wondering516 commented 1 month ago

Thanks for your great works, I am still confused about the part 4.4 in the paper.

Many Thanks for authors.

2hiTee commented 2 weeks ago

I have the same question, the author says first leverage 3DGS to reconstruct the object and render more views. But the paper says it just produce 16 views from the diffusion model, which I think it is not enough to run COLMAP to get the correct camera pose for 3DGS reconstruction.