# Seeing Small Faces from Robust Anchor's Perspective

arXiv: https://arxiv.org/abs/1802.09058 · HTML: https://mlsanity.com/paper/1802.09058/

## Key facts

- Cited by 11 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 Feb 2018.

## Built on by (most cited first, 11 of 11)

- [Towards Large-Scale Small Object Detection: Survey and Benchmarks](https://mlsanity.com/paper/2207.14096/) (2022)
- [Tracking Pedestrian Heads in Dense Crowd](https://mlsanity.com/paper/2103.13516/) (2021)
- [HLA-Face: Joint High-Low Adaptation for Low Light Face Detection](https://mlsanity.com/paper/2104.01984/) (2021)
- [MogFace: Towards a Deeper Appreciation on Face Detection](https://mlsanity.com/paper/2103.11139/) (2021)
- [CRFace: Confidence Ranker for Model-Agnostic Face Detection Refinement](https://mlsanity.com/paper/2103.07017/) (2021)
- [Detect Faces Efficiently: A Survey and Evaluations](https://mlsanity.com/paper/2112.01787/) (2021)
- [MANGO: Multimodal Attention-based Normalizing Flow Approach to Fusion Learning](https://mlsanity.com/paper/2508.10133/) (2025)
- [Probabilistic Ranking-Aware Ensembles for Enhanced Object Detections](https://mlsanity.com/paper/2105.03139/) (2021)
- [Wide Aspect Ratio Matching for Robust Face Detection](https://mlsanity.com/paper/2103.05993/) (2021)
- [Bounding-box deep calibration for high performance face detection](https://mlsanity.com/paper/2110.03892/) (2021)
- [Learning to Borrow Features for Improved Detection of Small Objects in Single-Shot Detectors](https://mlsanity.com/paper/2505.00044/) (2025)

## Abstract (arXiv)

This paper introduces a novel anchor design to support anchor-based face detection for superior scale-invariant performance, especially on tiny faces. To achieve this, we explicitly address the problem that anchor-based detectors drop performance drastically on faces with tiny sizes, e.g. less than 16x16 pixels. In this paper, we investigate why this is the case. We discover that current anchor design cannot guarantee high overlaps between tiny faces and anchor boxes, which increases the difficulty of training. The new Expected Max Overlapping (EMO) score is proposed which can theoretically explain the low overlapping issue and inspire several effective strategies of new anchor design leading to higher face overlaps, including anchor stride reduction with new network architectures, extra shifted anchors, and stochastic face shifting. Comprehensive experiments show that our proposed method significantly outperforms the baseline anchor-based detector, while consistently achieving state-of-the-art results on challenging face detection datasets with competitive runtime speed.