生成人脸
Generators-with-stylegan2
https://github.com/a312863063/generators-with-stylegan2
人脸生成器
图片动起来
Magic-animate
https://github.com/magic-research/magic-animate
AnimateAnyone
https://github.com/HumanAIGC/AnimateAnyone
图片卡通化
AnimeGAN
https://github.com/TachibanaYoshino/AnimeGAN
将图片和视频变成漫画风格
PhotoCartoon
https://github.com/minivision-ai/photo2cartoon
将照片卡通化
图片理解
MINI-GPT4
https://github.com/Vision-CAIR/MiniGPT-4
基于LLaMA2，理解图片
mPLUG-Owl
https://github.com/X-PLUG/mPLUG-Owl
CLIP
https://github.com/openai/CLIP
OpenAI开源CLIP模型（传入图片和问题），达摩院开源的CLIP模型：https://modelscope.cn/models/damo/cv_clip-interrogator/summary
BLIP-2
https://huggingface.co/docs/transformers/main/model_doc/blip-2
可用于在给定图像和可选文本提示的情况下生成文本。
LLaVA
https://github.com/haotian-liu/LLaVA
Demo网站：https://llava.hliu.cc/
图片生成
Fooocus
https://github.com/lllyasviel/Fooocus
无需复杂参数，专注prompt（不支持中文，ControlNet作者）
FaceChain
https://github.com/modelscope/facechain
开源的生成个人写真库facechain
Glyph-ByT5
https://github.com/AIGText/Glyph-ByT5
输入文字生成海报。
Disco Diffusion
https://github.com/alembics/disco-diffusion
Google Disco Diffusion，Colab: https://colab.research.google.com/github/alembics/disco-diffusion/blob/main/Disco_Diffusion.ipynb#scrollTo=CheckGPU
Gligen-GUI
https://github.com/mut-ex/gligen-gui
指定图片内各区域内容即可生成图片。
FLUX
FLUX模型是由 Black Forest Labs 开源的高质量文本到图像生成模型，它在多个维度上展现了卓越性能，尤其在文本引导的图像生成、多主体场景构建以及精细的手部细节生成等方面，实现了显著的提升，为文生图领域设定了新的技术标杆。
https://github.com/black-forest-labs/flux
一键换装
MiniBrush
https://huggingface.co/spaces/xichenhku/MimicBrush
OutfitAnyone
https://github.com/HumanAIGC/OutfitAnyone
https://github.com/ihmily/outfit-anyone
OOTDifusion
https://github.com/levihsu/OOTDiffusion
鉴别AI图片
StableSignature
https://github.com/facebookresearch/stable_signature
自动抠图
SegmentAnything
https://github.com/facebookresearch/segment-anything
RobustVideoMatting
https://github.com/PeterL1n/RobustVideoMatting
照片修复
CodeFormer
https://github.com/sczhou/CodeFormer
可视化工具
ComfyUI
Stable diffusion GUI
https://github.com/comfyanonymous/ComfyUI
数据标注
OpenMMLab
OpenMMLab: https://github.com/open-mmlab/playground/blob/main/README_zh-CN.md
OpenMMLab标注: https://github.com/open-mmlab/playground/blob/main/label_anything/readme_zh.md
OpenMMLab标注使用: https://mp.weixin.qq.com/s/02150wGJngAr_bZKNZUJLw
X-AnyLabeling
X-AnyLabeling数据标注：https://github.com/CVHub520/X-AnyLabeling/blob/main/README_zh-CN.md
模型微调
DreamBooth
DreamBooth微调图像模型: https://dreambooth.github.io/
Github: https://github.com/XavierXiao/Dreambooth-Stable-Diffusion
评论