lllyasviel/ControlNet

1---2license: openrail3---4 5This is the pretrained weights and some other detector weights of ControlNet.6 7See also: https://github.com/lllyasviel/ControlNet8 9# Description of Files10 11ControlNet/models/control_sd15_canny.pth12 13- The ControlNet+SD1.5 model to control SD using canny edge detection.14 15ControlNet/models/control_sd15_depth.pth16 17- The ControlNet+SD1.5 model to control SD using Midas depth estimation.18 19ControlNet/models/control_sd15_hed.pth20 21- The ControlNet+SD1.5 model to control SD using HED edge detection (soft edge).22 23ControlNet/models/control_sd15_mlsd.pth24 25- The ControlNet+SD1.5 model to control SD using M-LSD line detection (will also work with traditional Hough transform).26 27ControlNet/models/control_sd15_normal.pth28 29- The ControlNet+SD1.5 model to control SD using normal map. Best to use the normal map generated by that Gradio app. Other normal maps may also work as long as the direction is correct (left looks red, right looks blue, up looks green, down looks purple). 30 31ControlNet/models/control_sd15_openpose.pth32 33- The ControlNet+SD1.5 model to control SD using OpenPose pose detection. Directly manipulating pose skeleton should also work.34 35ControlNet/models/control_sd15_scribble.pth36 37- The ControlNet+SD1.5 model to control SD using human scribbles. The model is trained with boundary edges with very strong data augmentation to simulate boundary lines similar to that drawn by human.38 39ControlNet/models/control_sd15_seg.pth40 41- The ControlNet+SD1.5 model to control SD using semantic segmentation. The protocol is ADE20k.42 43ControlNet/annotator/ckpts/body_pose_model.pth44 45- Third-party model: Openpose’s pose detection model.46 47ControlNet/annotator/ckpts/hand_pose_model.pth48 49- Third-party model: Openpose’s hand detection model.50 51ControlNet/annotator/ckpts/dpt_hybrid-midas-501f0c75.pt52 53- Third-party model: Midas depth estimation model.54 55ControlNet/annotator/ckpts/mlsd_large_512_fp32.pth56 57- Third-party model: M-LSD detection model.58 59ControlNet/annotator/ckpts/mlsd_tiny_512_fp32.pth60 61- Third-party model: M-LSD’s another smaller detection model (we do not use this one).62 63ControlNet/annotator/ckpts/network-bsds500.pth64 65- Third-party model: HED boundary detection.66 67ControlNet/annotator/ckpts/upernet_global_small.pth68 69- Third-party model: Uniformer semantic segmentation.70 71ControlNet/training/fill50k.zip72 73- The data for our training tutorial.74 75# Related Resources76 77Special Thank to the great project - [Mikubill' A1111 Webui Plugin](https://github.com/Mikubill/sd-webui-controlnet) !78 79We also thank Hysts for making [Gradio](https://github.com/gradio-app/gradio) demo in [Hugging Face Space](https://huggingface.co/spaces/hysts/ControlNet) as well as more than 65 models in that amazing [Colab list](https://github.com/camenduru/controlnet-colab)! 80 81Thank haofanwang for making [ControlNet-for-Diffusers](https://github.com/haofanwang/ControlNet-for-Diffusers)!82 83We also thank all authors for making Controlnet DEMOs, including but not limited to [fffiloni](https://huggingface.co/spaces/fffiloni/ControlNet-Video), [other-model](https://huggingface.co/spaces/hysts/ControlNet-with-other-models), [ThereforeGames](https://github.com/AUTOMATIC1111/stable-diffusion-webui/discussions/7784), [RamAnanth1](https://huggingface.co/spaces/RamAnanth1/ControlNet), etc!84 85# Misuse, Malicious Use, and Out-of-Scope Use86 87The model should not be used to intentionally create or disseminate images that create hostile or alienating environments for people. This includes generating images that people would foreseeably find disturbing, distressing, or offensive; or content that propagates historical or current stereotypes.88 89