HoiM

Questions regarding blendshapes

1

Thank you for your great work. After reading the paper, I still have some questions. So I hope you or anyone else can answer me if possible. In the paper,...

NVRTC: How to implement cached_get_nvrtc_params in C++? Where is its overridden implementation in python?

1

When I was creating the TensorRT plugin using libspconv, I encountered that cached_get_nvrtc_params got called. I know the reason is that deps.is_nvrtc is true (in ConvTunerSimple_tune_and_cache.cc). And then the convolution...

Support for DeepCache

3

[DeepCache](https://huggingface.co/docs/diffusers/main/en/optimization/deepcache) is an optimization technique that reduces the computation of the UNet loop. I tested it with Stable Fast Diffusion and it is faster than Stable-Fast. It would be great...

What is the cost for training LADD?

2

Thank you for your excellent work! We are currently trying to apply LADD (Latent Adversarial Diffusion Distillation) on our own 2B text-to-image model. Can you provide any more experimental details,...

Inside this directory `VideoTetris/long/t2v_enhanced` with requirements installed and models downloaded, after running this command, ``` python inference_videotetris.py --num_frames 80 ``` I got: ``` /path/to/miniconda3/envs/st2v/lib/python3.10/site-packages/timm/models/layers/__init__.py:48: FutureWarning: Importing from timm.models.layers is deprecated,...