Octuplet Loss: Make Face Recognition Robust to Image Resolution
This addresses robustness issues in face recognition systems for applications like surveillance or mobile devices, but it is incremental as it builds on existing triplet loss and fine-tuning approaches.
The paper tackles the problem of face recognition performance degradation due to low-resolution images by proposing octuplet loss, a novel combination of triplet loss that fine-tunes models to improve cross-resolution verification. The method achieves 95.12% accuracy on the XQLFW dataset and 99.73% on LFW without harming high-resolution performance.
Image resolution, or in general, image quality, plays an essential role in the performance of today's face recognition systems. To address this problem, we propose a novel combination of the popular triplet loss to improve robustness against image resolution via fine-tuning of existing face recognition models. With octuplet loss, we leverage the relationship between high-resolution images and their synthetically down-sampled variants jointly with their identity labels. Fine-tuning several state-of-the-art approaches with our method proves that we can significantly boost performance for cross-resolution (high-to-low resolution) face verification on various datasets without meaningfully exacerbating the performance on high-to-high resolution images. Our method applied on the FaceTransformer network achieves 95.12% face verification accuracy on the challenging XQLFW dataset while reaching 99.73% on the LFW database. Moreover, the low-to-low face verification accuracy benefits from our method. We release our code to allow seamless integration of the octuplet loss into existing frameworks.