NVlabs / latentfusion

LatentFusion: End-to-End Differentiable Reconstruction and Rendering for Unseen Object Pose Estimation
https://arxiv.org/pdf/1912.00416.pdf
Other
213 stars 34 forks source link

Could you share the list of remaining model ids for your experiments and the pre-processed ShapeNet models? #2

Closed wangg12 closed 4 years ago

wangg12 commented 4 years ago

Hi authors, very nice work! I used your preprocess script for the ShapeNetCore.v2 data. However, after filtering out the models which are > 50MB and > 10 components, I can only get < 20k models. It is much fewer than what you reported in the paper (about 30k). So I wonder if you could share your list of model ids used in your experiments and the preprocessed ShapeNet models?

vtasStu commented 3 years ago

Hello, I use blender to deal with ShapeNetCore.v2 too. I use different version of Blender, and get different error. Which version of Blender do you use, and how do you download and install Blender?