Susceptibility to Image Resolution in Face Recognition and Trainings Strategies
This addresses a practical issue in face recognition for real-world applications where image resolutions vary, though the methods are incremental improvements on existing models.
The paper tackles the problem of face verification performance dropping significantly when images have mismatched resolutions, showing that accuracy can fall from 99.23% to 55% for very low-resolution images. It proposes training strategies, such as using mixed-resolution batches and siamese networks, which improve cross-resolution accuracy to around 70% and achieve an overall performance of 97.72%.
Face recognition approaches often rely on equal image resolution for verifying faces on two images. However, in practical applications, those image resolutions are usually not in the same range due to different image capture mechanisms or sources. In this work, we first analyze the impact of image resolutions on face verification performance with a state-of-the-art face recognition model. For images synthetically reduced to $5\,\times\,5$ px resolution, the verification performance drops from $99.23\%$ increasingly down to almost $55\%$. Especially for cross-resolution image pairs (one high- and one low-resolution image), the verification accuracy decreases even further. We investigate this behavior more in-depth by looking at the feature distances for every 2-image test pair. To tackle this problem, we propose the following two methods: 1) Train a state-of-the-art face-recognition model straightforwardly with $50\%$ low-resolution images directly within each batch. 2) Train a siamese-network structure and add a cosine distance feature loss between high- and low-resolution features. Both methods show an improvement for cross-resolution scenarios and can increase the accuracy at very low resolution to approximately $70\%$. However, a disadvantage is that a specific model needs to be trained for every resolution pair. Thus, we extend the aforementioned methods by training them with multiple image resolutions at once. The performances for particular testing image resolutions are slightly worse, but the advantage is that this model can be applied to arbitrary resolution images and achieves an overall better performance ($97.72\%$ compared to $96.86\%$). Due to the lack of a benchmark for arbitrary resolution images for the cross-resolution and equal-resolution task, we propose an evaluation protocol for five well-known datasets, focusing on high, mid, and low-resolution images.