Learning-Based Quality Assessment for Image Super-Resolution
This work provides a new, larger dataset and a better metric for researchers developing Super-Resolution algorithms, addressing a bottleneck in evaluating SR performance.
The authors address the lack of large-scale quality databases for Super-Resolution (SR) by creating SISAR, a database of 8,400 images with semi-automatic ratings. They then train DISQ, a two-stream deep neural network model, which outperforms existing SR quality metrics and shows good generalization.
Image Super-Resolution (SR) techniques improve visual quality by enhancing the spatial resolution of images. Quality evaluation metrics play a critical role in comparing and optimizing SR algorithms, but current metrics achieve only limited success, largely due to the lack of large-scale quality databases, which are essential for learning accurate and robust SR quality metrics. In this work, we first build a large-scale SR image database using a novel semi-automatic labeling approach, which allows us to label a large number of images with manageable human workload. The resulting SR Image quality database with Semi-Automatic Ratings (SISAR), so far the largest of SR-IQA database, contains 8,400 images of 100 natural scenes. We train an end-to-end Deep Image SR Quality (DISQ) model by employing two-stream Deep Neural Networks (DNNs) for feature extraction, followed by a feature fusion network for quality prediction. Experimental results demonstrate that the proposed method outperforms state-of-the-art metrics and achieves promising generalization performance in cross-database tests. The SISAR database and DISQ model will be made publicly available to facilitate reproducible research.