Abstract
This study proposes a novel face detector called DEFace that focuses on the challenging tasks of face detection to cope with a small size that is under 12 pixels and occlusions due to a mask or human body parts. This study proposed the extended feature pyramid network (FPN) module to detect small faces by expanding the range of P layer, and the network by adding a receptive context module (RCM) after each predicted feature head from the top-down pathway in the FPN architecture to enhance the feature discriminability and the robustness. Based on the FPN principle, the combination between the low- and high-resolutions are beneficial for object detection with different object sizes. Furthermore, with assistance from the RCM, the proposed method can use a broad range of context information especially for small faces. To evaluate the performance of the proposed method, various public face datasets are used such as the WIDER Face dataset, the face detection dataset and benchmark (FDDB), and the masked faces (MAFA) dataset, which consist of challenging samples such as small face regions and occlusions by hair or other people. The results indicate that DEFace can detect the face region more accurately in comparison to the other state-of-the-art methods while maintaining the processing time.
Highlights
Face detection has been extensively studied for decades
This study proposes a novel densely connected network for face detection, especially well-behaved for small faces whose bounding box sizes are smaller than 12 pixels
There are three contributions in this paper as follows: 1) We propose a novel deep network model by adapting the RetinaNet [10] structure with extended feature pyramid network (FPN) module and additional context module block that can strengthen the feature maps and assist the model to cope with examples of faces such as small faces in the WIDER Face dataset [1], the occluded faces in the face detection dataset and benchmark (FDDB) dataset [2], and the masked faces (MAFA) dataset [3], [38]
Summary
Face detection has been extensively studied for decades. Inherit from recent significant achievements from deep learning-based methods for object detection [4]–[10], face detection has recently obtained noticeable progress [11]–[16]. The most recent state-of-the-art deep learning-based methods [11], [17]–[20] for object detection, or face in particular, can be classified into two groups: single-stage methods [5], [10], [21] and two-stage methods [4], [22]. The single-stage method densely samples face locations and the scales with feature pyramids [9]. This demonstrates a balance between the accuracy and the speed processing time in comparison to other counterpart two-stage detection methods [4], [22].
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
