Unverified60% confidenceFactExact time
ArcFace was proposed by Deng Jiankang et al. in 2019 and employs an additive angular margin loss.
2
Sources
60%
Confidence
Long-term
Relevance
8/7/2026
First Seen
Sources
Related Claims
UnverifiedArcFace proposed the Additive Angular Margin Loss in 2018, achieving over 99.8% recognition accuracy on standard benchmarks like LFW77% similarUnverifiedThe core idea of loss functions such as ArcFace and CosFace is to artificially enlarge inter-class margins and compress intra-class distributions in angular space to improve discriminability.73% similarVerifiedArcFace is a deep metric learning method optimized specifically for face recognition that outperforms general visual encoders in face identity feature extraction accuracy by introducing an additive angular margin loss function.72% similarUnverifiedArcFace first applies L2 normalization to both the feature embedding vectors and the classification weight matrix, then adds a fixed angular margin (typically 0.5 radians) to the cosine of the angle, scales the result, and feeds it into Softmax to compute the loss.69% similarUnverifiedArcReel's core design goal is to solve the problem of character and scene consistency across shots.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/705629API
curl https://kongchang.com/api/v1/knowledge/claims/705629MCP
get_claim(id=705629)