First of all, thank the author for his amazing work. I have been able to train and generate a color 3D model, but all of this is based on the data provided in the article. I rewrite test.py to load the pre-trained model to reconstruct the color 3D model from a single external image. But the effect is not ideal. It seems that shape reconstruction cannot be completed, and texture learning is currently no problem.
Below is my test_generate.py file
test_generate.py
Please advise, thank you very much, and look forward to your reply
First of all, thank the author for his amazing work. I have been able to train and generate a color 3D model, but all of this is based on the data provided in the article. I rewrite test.py to load the pre-trained model to reconstruct the color 3D model from a single external image. But the effect is not ideal. It seems that shape reconstruction cannot be completed, and texture learning is currently no problem. Below is my test_generate.py file test_generate.py Please advise, thank you very much, and look forward to your reply