search

hansen7 / OcCo

[ICCV '21] "Unsupervised Point Cloud Pre-training via Occlusion Completion"
https://hansen7.github.io/OcCo/
MIT License
233 stars 26 forks source link

two types of pre-trained weights #28

Closed zshyang closed 2 years ago

zshyang commented 2 years ago

Hi @hansen7 I saw you have released two types of weights. They are _cls.pth and _seg.pth. My guess is that they have the same encoder but just the seg one has a randomly initialized segmentation network appended to the pre-trained encoder. Not sure if my guess is correct or not so need your information. Best regards.

hansen7 commented 2 years ago

Hi @zshyang, it is :)

zshyang commented 2 years ago

Hi @hansen7 , thanks for the reply. I am still curious about why do you have such an implementation. Why do not just save the *_cls.pth and load it to the encoder part in the segmentation tasks.

hansen7 commented 2 years ago

Oh, the encoders are slightly different, for classification and segmentation. You can check the original papers and codes :)

zshyang commented 2 years ago

thanks!