# HAMBox: Delving into Online High-quality Anchors Mining for Detecting Outer Faces

Source: [HAMBox: Delving into Online High-quality Anchors Mining for Detecting Outer Faces](<https://www.sota2.com/research/paper/68fa66e61e209131cea12f6d>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-25T22:01:21.007Z

Authors: Yang Liu, Xu Tang, Xiang Wu, Junyu Han, Jingtuo Liu, Errui Ding

Published: 2019-12-19T14:44:26.000Z

## Abstract

Current face detectors utilize anchors to frame a multi-task learning problem which combines classification and bounding box regression. Effective anchor design and anchor matching strategy enable face detectors to localize faces under large pose and scale variations. However, we observe that more than 80% correctly predicted bounding boxes are regressed from the unmatched anchors (the IoUs between anchors and target faces are lower than a threshold) in the inference phase. It indicates that these unmatched anchors perform excellent regression ability, but the existing methods neglect to learn from them. In this paper, we propose an Online High-quality Anchor Mining Strategy (HAMBox), which explicitly helps outer faces compensate with high-quality anchors. Our proposed HAMBox method could be a general strategy for anchor-based single-stage face detection. Experiments on various datasets, including WIDER FACE, FDDB, AFW and PASCAL Face, demonstrate the superiority of the proposed method. Furthermore, our team win the championship on the Face Detection test track of WIDER Face and Pedestrian Challenge 2019. We will release the codes with PaddlePaddle.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Face Detection on WIDER FACE 1 (val)](<https://www.sota2.com/research/sota/face-detection-on-wider-face-1-val.md>): Face Detection; WIDER FACE 1 (val); Detection Score (Easy): 95.27.

- [Face Detection on WIDER FACE (val)](<https://www.sota2.com/research/sota/face-detection-on-wider-face-val.md>): Face Detection; WIDER FACE (val); mAP (Easy): 95.27.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/1912.09231>)

[PDF](<https://arxiv.org/pdf/1912.09231.pdf>)