Friday

I also want to learn about more details about Image-to-3D generation! I finetune Zero123 with some images and get the improvement on novel view synthesis. However, the results are quite...

same issue here, but I use sd model v2.1. The wired thing is that I use inject_trainable_lora into sd model with target_instead_model=["CrossAttention"] and return no parameters. For more details, my...

Hi, everyone! I just find my issue is by caused by data_type!!!! My code is as follow and I hope this can help any of you meet the same problem....

> I would also like to share some insights on Betas. Accidentally i visualized some mocap data with only the first Betas (1st of 10 parameters) correct. The overlay is...

Same issue here. Look forward to new ckpt

@ShuyUSTC Hi, did you run the updated code? However, Issue still exists on my own data. So strange,,,

> > @ShuyUSTC Hi, did you run the updated code? However, Issue still exists on my own data. So strange,,, > > Are you using this new [checkpoint](https://huggingface.co/ttxskk/AiOS/tree/main)? If not,...

I try to use TalkSHOW code to visualize data but I get the bad result.  Do you know the reason? My code is as follow (from `TalkSHOW/scripts/demo.py`): ```python lower_pose...

> Owner > Hi James, you may want to pay attention to the code here: > > https://github.com/JeremyCJM/DiffSHEG/blob/3ebf3058f48cba3da9146afb7623e9ec1ab9e9a5/datasets/show.py#L146 > > . The order of channels for pose should be carefully...