Divano
Divano
Hello,I meeting a problem as follows,I kept all the input formats consistent with the author. I don't know what could be a bug.I am new at it,if you give me...
Hello,could you write the Dependencies in readme?eg. I install TF2 but the code need TF1.I would be very grateful if you could accept my opinion.Thanks.
hello!nice work! I want to know if this work could use depth as a control,assume there already had mutil-view depth images rendered from a mesh.
### Describe the bug diffusers0.19.3，in pipeline_controlnet.py， in \__call__(),`cross_attention_kwargs: Optional[Dict[str, Any]] = None,`, but pipe(cross_attention_kwargs={"scale": 0.8}) has problem: File "/lib/python3.8/site-packages/diffusers/models/attention_processor.py", line 322, in forward return self.processor( TypeError: __call__() got an unexpected...
At the end of the generate_texture.py, there is a post-process operation, i want to know what's the use of it. ` # post-process if args.post_process: del controlnet del ddim_sampler inpainting...
in scripts/test_depth_fix_frames.sh，line 21, --ckpt_path weights/depth_gen_new.pth. but after download [Depth-conditioned generation model](https://www.dropbox.com/scl/fi/56hcmoj0tx7lza7s2m0jq/depth_gen.ckpt?rlkey=upcdbd4kxd9zwms78dssm3gh7&dl=0) the name is depth_gen.ckpt ,mismatch name in test_depth_fix_frames.sh, is that a bug?
Has controlnet been developed for SV3D so far? Especially depth control?
Is the point cloud encoder universal, or does it need to be retrained for each category of point clouds?
Do you train point cloud autoencoders for each individual category, or do you train them for all categories in one go?