Prathamesh Dinkar

Using a dictionary to store the mesh data instead of numpy. The numpy requires all the data to be in the same shape which is not true for "ve", so...

For training models other than humans, I think a custom SMPL is needed which can generate deformed shapes of your desired model (which is very difficult to find). And once...

The results provided in the paper are not matching. I'm getting the following results after training using the train script and evaluated using the given script: 

Can you explain for can we implement the Mahalanobis in semantic segmentation for OOD detection?

How to convert given intrinsic data (Faces and edge lengths) to extrinsic data like an OBJ file?

How to use the output files to create a OBJ, do you know?

Can you please tell me how to save the Delaunay refined mesh as a OBJ file?

The model output for the semantic predictions is in the format: semantic logits : (batch, 81, 81, num_classes) (For a crop_size of 321) semantic scores : (batch, img_height, img_width, num_classes)....

Encoder instead of Decoder

2

You have used a decoder architecture, what do you think about using an encoder instead?