PyramidBox++: High Performance Detector for Finding Tiny Face

TL;DR

PyramidBox++ is a high-performance face detector that improves upon previous methods by introducing advanced data augmentation, dual pyramid anchors, and a dense context module, achieving state-of-the-art results on challenging datasets.

Contribution

The paper presents PyramidBox++ with novel techniques like Balanced-data-anchor-sampling, Dual-PyramidAnchors, and Dense Context Module, enhancing face detection especially for tiny faces.

Findings

Achieves state-of-the-art performance on WIDER FACE dataset.

Improves detection of tiny faces with new modules.

Enhances feature learning and receptive field size.

Abstract

With the rapid development of deep convolutional neural network, face detection has made great progress in recent years. WIDER FACE dataset, as a main benchmark, contributes greatly to this area. A large amount of methods have been put forward where PyramidBox designs an effective data augmentation strategy (Data-anchor-sampling) and context-based module for face detector. In this report, we improve each part to further boost the performance, including Balanced-data-anchor-sampling, Dual-PyramidAnchors and Dense Context Module. Specifically, Balanced-data-anchor-sampling obtains more uniform sampling of faces with different sizes. Dual-PyramidAnchors facilitate feature learning by introducing progressive anchor loss. Dense Context Module with dense connection not only enlarges receptive filed, but also passes information efficiently. Integrating these techniques, PyramidBox++ is constructed and achieves state-of-the-art performance in hard set.