FDLite: A Single Stage Lightweight Face Detector Network

Yogesh Aggarwal,Prithwijit Guha

2024-06-27

Abstract:Face detection is frequently attempted by using heavy pre-trained backbone networks like ResNet-50/101/152 and VGG16/19. Few recent works have also proposed lightweight detectors with customized backbones, novel loss functions and efficient training strategies. The novelty of this work lies in the design of a lightweight detector while training with only the commonly used loss functions and learning strategies. The proposed face detector grossly follows the established RetinaFace architecture. The first contribution of this work is the design of a customized lightweight backbone network (BLite) having 0.167M parameters with 0.52 GFLOPs. The second contribution is the use of two independent multi-task losses. The proposed lightweight face detector (FDLite) has 0.26M parameters with 0.94 GFLOPs. The network is trained on the WIDER FACE dataset. FDLite is observed to achieve 92.3\%, 89.8\%, and 82.2\% Average Precision (AP) on the easy, medium, and hard subsets of the WIDER FACE validation dataset, respectively.

Computer Vision and Pattern Recognition

What problem does this paper attempt to address?

The paper aims to address the demand for lightweight face detectors on edge devices in real-time applications. Specifically, the goal of the paper is to reduce the number of floating-point operations in the network without sacrificing accuracy, thereby designing a face detection system that can operate at low computational cost.

To achieve this goal, the paper proposes a single-stage lightweight face detection network named FDLite. The main contributions include:
1. Designing a customized lightweight backbone network (BLite) with 0.167M parameters and 0.52 GFLOPs.
2. Proposing two separate multi-task loss functions.
3. The entire face detection network FDLite has 0.26M parameters and 0.94 GFLOPs, and achieves average precision (AP) of 92.3%, 89.8%, and 82.2% on the easy, medium, and hard subsets of the WIDER FACE validation dataset, respectively.

With these improvements, FDLite achieves performance comparable to or even better than existing state-of-the-art methods while maintaining a low computational cost.