Deep Tiny Network for Recognition-Oriented Face Image Quality Assessment
This work addresses face recognition instability for scenarios with intra-variations in images, offering an efficient solution, though it appears incremental as it builds on existing quality assessment methods.
The authors tackled the problem of face recognition instability caused by low-quality face images by proposing a non-reference image quality assessment method that directly links image quality to recognition performance, resulting in a deep Tiny Face Quality network that outperforms state-of-the-art methods in effectiveness and efficiency on benchmarks like IJB-B and YTF.
Face recognition has made significant progress in recent years due to deep convolutional neural networks (CNN). In many face recognition (FR) scenarios, face images are acquired from a sequence with huge intra-variations. These intra-variations, which are mainly affected by the low-quality face images, cause instability of recognition performance. Previous works have focused on ad-hoc methods to select frames from a video or use face image quality assessment (FIQA) methods, which consider only a particular or combination of several distortions. In this work, we present an efficient non-reference image quality assessment for FR that directly links image quality assessment (IQA) and FR. More specifically, we propose a new measurement to evaluate image quality without any reference. Based on the proposed quality measurement, we propose a deep Tiny Face Quality network (tinyFQnet) to learn a quality prediction function from data. We evaluate the proposed method for different powerful FR models on two classical video-based (or template-based) benchmark: IJB-B and YTF. Extensive experiments show that, although the tinyFQnet is much smaller than the others, the proposed method outperforms state-of-the-art quality assessment methods in terms of effectiveness and efficiency.