EResFD: Rediscovery of the Effectiveness of Standard Convolution for Lightweight Face Detection
This work addresses efficiency challenges in face detection for mobile or embedded systems, offering a novel architectural improvement that is incremental but provides specific gains.
The paper tackles the problem of efficient face detection by re-evaluating standard convolution as a lightweight backbone, showing that heavily channel-pruned standard convolutions outperform depthwise separable convolutions in accuracy and speed with similar parameters, achieving 80.4% mAP on WIDER FACE Hard subset with 37.7 ms inference time on CPU.
This paper analyzes the design choices of face detection architecture that improve efficiency of computation cost and accuracy. Specifically, we re-examine the effectiveness of the standard convolutional block as a lightweight backbone architecture for face detection. Unlike the current tendency of lightweight architecture design, which heavily utilizes depthwise separable convolution layers, we show that heavily channel-pruned standard convolution layers can achieve better accuracy and inference speed when using a similar parameter size. This observation is supported by the analyses concerning the characteristics of the target data domain, faces. Based on our observation, we propose to employ ResNet with a highly reduced channel, which surprisingly allows high efficiency compared to other mobile-friendly networks (e.g., MobileNetV1, V2, V3). From the extensive experiments, we show that the proposed backbone can replace that of the state-of-the-art face detector with a faster inference speed. Also, we further propose a new feature aggregation method to maximize the detection performance. Our proposed detector EResFD obtained 80.4% mAP on WIDER FACE Hard subset which only takes 37.7 ms for VGA image inference on CPU. Code is available at https://github.com/clovaai/EResFD.