AnyFace: A Data-Centric Approach For Input-Agnostic Face Detection

Abstract

<p>Face detection is a mandatory step in many computer vision applications, such as face recognition, emotion recognition, age detection, virtual makeup, and vital sign monitoring. Thanks to advancements in deep learning and the introduction of annotated large-scale datasets, numerous applications have been developed for human faces. Recently, other domains, such as animals and cartoon characters, have started gaining attention but still lag far behind human faces. The biggest challenge is the limited number of annotated face datasets in these domains. The manual labeling of large-scale datasets is tedious and requires substantial human labor. In this regard, we present an input-agnostic face detector to ease the annotation of various face datasets. We propose a simple but effective data-centric approach instead of building a specific neural network architecture. Specifically, we trained a face detection model, YOLO5Face, on human, animal, and cartoon face datasets. The experiments show that the model can achieve accurate results in all domains. In addition, the model achieved decent results for animals and cartoon characters different from the ones in the training set. This implies that the model can extract agnostic facial features. We have made the source code and pre-trained models publicly available at https://github.com/IS2AI/AnyFace to stimulate research in these fields.</p>

A Lightweight Face Detector via Bi-Stream Convolutional Neural Network and Vision Transformer

“…At first, the DP layer splits X 2 into K = H/2 * W/2 patches. The dimension of each patch is R 2×2×T ; P k denotes the kth patch In Equation (2). Here, R(•) represents the reorganization of the tensor at (n, m) of each patch into a new tensor X i 2 ∈ R h×w×T , i ∈ {1, 2, 3, 4}, where the correspondence among i , m, and n can be described as The proposed DP layer effectively reduces the loss of low-level features by transferring the information of adjacent pixels from the spatial dimension to the channel dimension.…”

“…The process of facial detection involves the meticulous identification and localization of human facial features within images. Face detection is a crucial task in the field of computer vision and has garnered significant attention due to its pivotal role in downstream applications like face recognition and reconstruction [1][2][3][4]. In recent years, face detection methods [5,6] have witnessed significant advancements in detection accuracy and speed owing to the emergence and refinement of convolutional neural networks (CNNs).…”