optas / shapeglot Goto Github PK

Thanks for releasing the project!

However, I have faced some errors when running the notebook for training the network, as shown below:

It seems that the data type for the text_padded stored is not correct, which was generated directly using another notebook provided. Any hints for resolving the issues are highly appreciated!

Good Morning,
I am writing to ask if it is possible to get the image-based (VGG-16) and point-cloud-based (PC-AE) pretrained features for the Shapenet category of tables. From your repository, only the features for the chairs can be downloaded.
In case this is not possible, I would really appreciate if you could :

• provide the details about how you captured the renderings of the shapes
• provide the pretrained weights of the PC Encoder and VGG-16 you have used.

Regards,
Andrea

Where do I find the Point cloud dataset for each shape. The dataset link given in the form doesn't contain the PC data