EfficientFace: an efficient deep network with feature enhancement for accurate face detection
Guangtao Wang,Jun Li,Zhijian Wu,Jianhua Xu,Jifeng Shen,Wankou Yang
DOI: https://doi.org/10.1007/s00530-023-01134-6
IF: 3.9
2023-07-14
Multimedia Systems
Abstract:In recent years, deep convolutional neural networks (CNN) have significantly advanced face detection. In particular, lightweight CNN-based architectures have achieved great success due to their low-complexity structure facilitating real-time detection tasks. However, current lightweight CNN-based face detectors trading accuracy for efficiency have inadequate capability in handling insufficient feature representation, faces with unbalanced aspect ratios and occlusion. Consequently, they exhibit deteriorated performance far lagging behind the deep heavy detectors. To achieve efficient face detection without sacrificing accuracy, we design an efficient deep face detector termed EfficientFace in this study, which contains three modules for feature enhancement. To begin with, we design a novel cross-scale feature fusion strategy to facilitate bottom-up information propagation, such that fusing low-level and high-level features is further strengthened. Besides, this is conducive to estimating the locations of faces and enhancing the descriptive power of face features. Second, we introduce a Receptive Field Enhancement module to consider faces with various aspect ratios. Third, we add an Attention Mechanism module for improving the representational capability of occluded faces. We have evaluated EfficientFace on four public benchmarks and experimental results demonstrate the appealing performance of our method. In particular, our model respectively achieves 95.1% (Easy), 94.0% (Medium) and 90.1% (Hard) on a validation set of WIDER Face dataset, which is competitive with heavyweight models with only 1/15 computational costs of the state-of-the-art MogFace detector.
computer science, information systems, theory & methods