Unverified60% confidenceFactExact time
RetinaFace is built on a Feature Pyramid Network (FPN) to construct multi-scale feature maps and can detect extremely small faces as small as 16×16 pixels.
2
Sources
60%
Confidence
Long-term
Relevance
8/8/2026
First Seen
Sources
Related Claims
UnverifiedRetinaFace在WIDER FACE数据集的Hard子集上达到了当时的最优性能75% similarUnverifiedRetinaFace pushed face detection accuracy to new heights in 2019, maintaining robustness under extreme poses and occlusion conditions75% similarUnverifiedThe research team re-encoded the fisheye and half equirectangular projection footage using x265, covering 24 clips, both eyes, four bitrate points, and nine viewing directions.71% similarUnverified在Retina屏幕上进行坐标映射时需额外乘以设备像素比（DPR，通常为2），否则所有点击都会落在屏幕左上角四分之一区域内65% similarUnverifiedThe TrueDepth camera module that Face ID relies on consists of multiple components such as an infrared lens, a flood illuminator, and a dot projector; it needs to occupy a relatively fixed horizontal strip at the top of the screen and maintain an imaging distance of about 25–50 centimeters from the user's face.65% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/705823API
curl https://kongchang.com/api/v1/knowledge/claims/705823MCP
get_claim(id=705823)