人脸属性模型FairFace MogFace为当前SOTA的人脸检测方法,已在Wider Face六项榜单上霸榜一年以上,后续被CVPR2022录取(论文地址,代码地址),该方法的主要贡献是从下面三个角度提升人脸检测器: MogFace在WiderFace榜单上的指标如下: ![模型效果]() 本模型可以检测输入图片中人的性别和年龄范围:[0-2, 3-9, 10-19, 20-29, 30-39, 40-49, 50-59, 60-69, 70+]。 测试时主要的预处理如下: 本模型及代码来自开源社区(地址),请遵守相关许可。 如果你觉得这个该模型对有所帮助,请考虑引用下面的相关的论文:ArcFace 模型介绍
模型描述
模型效果
模型使用方式和使用范围
代码范例
from modelscope.pipelies import pipelie
from modelscope.utils.costat import Tasks
mog_face_detectio_fuc = pipelie(Tasks.face_detectio, 'damo/cv_reset101_face-detectio_cvpr22papermogface')
src_img_path = 'https://modelscope.oss-c-beijig.aliyucs.com/test/images/mog_face_detectio.jpg'
raw_result = mog_face_detectio_fuc(src_img_path)
prit('face detectio output: {}.'.format(raw_result))
# if you wat to show the result, you ca ru
from modelscope.utils.cv.image_utils import draw_face_detectio_o_lm_result
from modelscope.preprocessors.image import LoadImage
import cv2
import umpy as p
# load image from url as rgb order
src_img = LoadImage.covert_to_darray(src_img_path)
# save src image as bgr order to local
src_img = cv2.cvtColor(p.asarray(src_img), cv2.COLOR_RGB2BGR)
cv2.imwrite('src_img.jpg', src_img)
# draw dst image from local src image as bgr order
dst_img = draw_face_detectio_o_lm_result('src_img.jpg', raw_result)
# save dst image as bgr order to local
cv2.imwrite('dst_img.jpg', dst_img)
# show dst image by rgb order
import matplotlib.pyplot as plt
dst_img = cv2.cvtColor(p.asarray(dst_img), cv2.COLOR_BGR2RGB)
plt.imshow(dst_img)
使用方式
目标场景
模型局限性及可能偏差
预处理
来源说明
引用
@article{karkkaie2019fairface,
title={Fairface: Face attribute dataset for balaced race, geder, ad age},
author={K{\"a}rkk{\"a}ie, Kimmo ad Joo, Jugseock},
joural={arXiv preprit arXiv:1908.04913},
year={2019}
}
点击空白处退出提示
评论