使用改进的 BiSeNet 在 PyTorch 中进行人脸解析
face-segmentation 开源项目推荐(39 个)
face-segmentation 话题在 GitHub 上没有填写简介,这里按 Star 数汇总最相关的热门仓库。
热门项目 前 20
按 Star 数排序展示前 20 个 · GitHub 共 39 个仓库一个用于人脸解析、识别、生成和编辑的大规模人脸数据集。
面向 Python 的统一人脸分析库 | 在一个 API 下提供检测、对齐、关键点、人脸网格、识别、解析、视线、属性和防欺骗。
使用 BiSeNet 的实时人脸解析和面部语义分割——PyTorch 训练、ONNX 导出、预训练权重。
一组移植到 Keras 用于人脸分析的深度学习框架。
[CVPRW 2022] Delving into High-Quality Synthetic Face Occlusion Segmentation Datasets
Code repository for "Interpretable and Accurate Fine-grained Recognition via Region Grouping", CVPR 2020 (Oral)
Reliable face detection, segmentation and enhancement for ComfyUI with custom-trained models that work across anime, realism, extreme poses and occlusion.
Face/Hair segmentation images dataset
Face segmentation with CNN and CRF
Towards Learning Structure via Consensus for Face Segmentation and Parsing (CVPR 2020)
该项目在 GitHub 上没有填写描述
使用语义分割解析面部的不同部分。
ComfyUI custom nodes for model-aware single- and multi-face repair using crop-based img2img, semantic masks, color matching, and seamless blending for realistic, anime, and illustrated images.
Hair Color Change Using Pytorch Model
Crop out (and optionally remove background and correct roll of) faces in an image. Implements a custom pipeline using Mediapipe's FaceDetection and FaceMesh networks
Developing code on semantic segmentation for Extended Labeled Faces in the Wild
SuperGAN aims to develope subject agnostic real-time Face Swaping.
[IEEE T-BIOM] FaceXBench: Evaluating Multimodal LLMs on Face Understanding
The MFSD (Masked Face Segmentation Dataset) is a comprehensive dataset designed to advance research in masked face related tasks such as segmentation. This dataset is especially relevant in the context of the COVID-19 pandemic, where mask-wearing has become widespread.
关于这个话题
- 清单取自 GitHub 话题页并按 Star 数排序,每次展示前 20 个项目;原始页面见 https://github.com/topics/face-segmentation。
- 该话题在 GitHub 上共有 39 个仓库, 本站收录的是其中最热门的 20 个,不是全量列表。
- Star / Fork 每日自动刷新一次,实时数据以 GitHub 为准。
- 当前 20 个项目累计 8548 Star,覆盖 2 种主要语言,数据更新于 2026-11-07。
- 中文介绍来自 GitHub 官方话题说明(github/explore)的机器翻译,可点「查看原文」对照;译文缺失时会先展示原文。