ReconX

ReconX copied to clipboard

Hi, thanks for the wonderful work. I have a questions about point cloud embedder: DUSt3R produces point clouds for each pair of images. What is the one that is fed...

junyiouy

May I ask when it can be used?

1

After reading your introduction, I am looking forward to it. I am currently researching the use of 3DGS to generate Gaussian models for AI videos, which can then be used...

DreamerYinYu

Hello, @liuff19, Thanks for your great work! I have a question about why ReconX chooses DynamiCrafter as the video diffusion backbone. In fact, there are many alternative video diffusion backbone,...

wzy-99

Thanks for your good work! I have some questions about the point cloud embedder shown in Figure2. There is almost no information about the point cloud embedder (Dust3R dose not...

ewrfcas

Questions about equations in the paper.

1

Hello, @liuff19 Recently, I found your project very interesting and started reading your paper. However, I have some questions on some equations in the paper. 1. For equations (12) $$\...

frankkim1108

Opensource Timeline

7

Hi @liuff19, Thanks for your excellent work! It's so fascinating and it's my favorite work in 2024 until now! I was wondering what's the estimated timeline for opensource? Thanks in...

rwn17

How to implement the LPIPS loss? I tested this loss but the performance was bad, especially the PSNR decreased significantly. Thanks.

whu-lyh