ONNX 模型库
返回模型

说明文档

来自 insightface 项目 的 ONNX 模型。

如何使用

pip install dghs-realutils>=0.1.0
from realutils.face.insightface import isf_face_batch_similarity, isf_analysis_faces, isf_faces_visualize

image_path = "/your/image/file"
# 获取所有面部分析结果
faces = isf_analysis_faces(image_path)
print(faces)

# 比较它们
print(isf_face_batch_similarity([face.embedding for face in faces]))

# 可视化
isf_faces_visualize(image_path, faces).show()

可用模型

我们在一些人脸识别评估数据集上评估了所有这些模型。

  • CFPW(500 个身份/7K 张图像/7K 对)[1]
  • LFW(5749 个身份/13233 张图像/6K 对)[2]
  • CALFW(5749 个身份/13233 张图像/6K 对)[3]
  • CPLFW(5749 个身份/13233 张图像/6K 对)[4]

以下是完整结果和推荐阈值。

  • Det:人脸检测和关键点定位的成功率。
  • Rec-F1:人脸识别中达到的最大 F1 分数。
  • Rec-Thresh:由最大 F1 分数确定的最佳阈值。
Model Eval ALL (Det/Rec-F1/Rec-Thresh) Eval CALFW (Det/Rec-F1/Rec-Thresh) Eval CFPW (Det/Rec-F1/Rec-Thresh) Eval CPLFW (Det/Rec-F1/Rec-Thresh) Eval LFW (Det/Rec-F1/Rec-Thresh)
buffalo_l 99.88% / 98.34% / 0.2203 100.00% / 95.75% / 0.2273 99.99% / 99.66% / 0.1866 99.48% / 96.41% / 0.2207 100.00% / 99.85% / 0.2469
buffalo_s 99.49% / 96.87% / 0.1994 99.99% / 94.45% / 0.2124 99.65% / 98.64% / 0.1845 98.04% / 92.61% / 0.2019 100.00% / 99.68% / 0.2314

[1] Sengupta Soumyadip, Chen Jun-Cheng, Castillo Carlos, Patel Vishal M, Chellappa Rama, Jacobs David W, Frontal to profile face verification in the wild, WACV, 2016.

[2] Gary B. Huang, Manu Ramesh, Tamara Berg, and Erik Learned-Miller. Labeled Faces in the Wild: A Database for Studying Face Recognition in Unconstrained Environments, 2007.

[3] Zheng Tianyue, Deng Weihong, Hu Jiani, Cross-age lfw: A database for studying cross-age face recognition in unconstrained environments, arXiv:1708.08197, 2017.

[4] Zheng, Tianyue, and Weihong Deng. Cross-Pose LFW: A Database for Studying Cross-Pose Face Recognition in Unconstrained Environments, 2018.

deepghs/insightface

作者 deepghs

feature-extraction
↓ 0 ♥ 9

创建时间: 2025-02-23 16:02:30+00:00

更新时间: 2025-02-25 16:30:44+00:00

在 Hugging Face 上查看

文件 (44)

.gitattributes
README.md
buffalo_l/1k3d68.onnx ONNX
buffalo_l/2d106det.onnx ONNX
buffalo_l/CALFW_metrics.json
buffalo_l/CALFW_plot.png
buffalo_l/CALFW_sims.csv
buffalo_l/CFPW_metrics.json
buffalo_l/CFPW_plot.png
buffalo_l/CFPW_sims.csv
buffalo_l/CPLFW_metrics.json
buffalo_l/CPLFW_plot.png
buffalo_l/CPLFW_sims.csv
buffalo_l/LFW_metrics.json
buffalo_l/LFW_plot.png
buffalo_l/LFW_sims.csv
buffalo_l/det_10g.onnx ONNX
buffalo_l/genderage.onnx ONNX
buffalo_l/metrics.json
buffalo_l/plot.png
buffalo_l/ref.json
buffalo_l/sims.csv
buffalo_l/w600k_r50.onnx ONNX
buffalo_s/1k3d68.onnx ONNX
buffalo_s/2d106det.onnx ONNX
buffalo_s/CALFW_metrics.json
buffalo_s/CALFW_plot.png
buffalo_s/CALFW_sims.csv
buffalo_s/CFPW_metrics.json
buffalo_s/CFPW_plot.png
buffalo_s/CFPW_sims.csv
buffalo_s/CPLFW_metrics.json
buffalo_s/CPLFW_plot.png
buffalo_s/CPLFW_sims.csv
buffalo_s/LFW_metrics.json
buffalo_s/LFW_plot.png
buffalo_s/LFW_sims.csv
buffalo_s/det_500m.onnx ONNX
buffalo_s/genderage.onnx ONNX
buffalo_s/metrics.json
buffalo_s/plot.png
buffalo_s/ref.json
buffalo_s/sims.csv
buffalo_s/w600k_mbf.onnx ONNX