Shikun Liu

Hi @ChandanVerma, thanks for the reach out. Fine-tuning on the custom dataset is very straightforward. You just need to 1) Prepare your .json data list file, similar to what I...

Please re-raise an issue when you have other more concrete questions.

Wondering too. These values also seem to be different from different example files...

I didn't realise there was a gym in Isaac Sim as well. But I do know they recently release another Gym-like library called Orbit which is built on top of...

Hi @fradif96, the network is fine-tuned from the StableDiffusion v1.5 checkpoints.

1. CLIP can accept multiple reference images. But since CLIP is trained to extract semantics, frozen CLIP mainly produces semantic information. That's the main motivation that Zero-1-to-3 and other methods...

No. Because they all have white background.

Of course. EscherNet with scene-level data is one of the most important future directions that we are currently exploring. Just simply obtaining these data is not that straightforward.