Learning Single-Image Super-Resolution in the JPEG Compressed Domain
This addresses efficiency issues for researchers and practitioners in computer vision, though it is incremental as it extends prior work from recognition to restoration tasks.
The paper tackles the data loading bottleneck in deep learning by training models directly on JPEG features for single-image super-resolution, achieving a 2.6x speedup in data loading and a 2.5x speedup in training while maintaining comparable visual quality.
Deep learning models have grown increasingly complex, with input data sizes scaling accordingly. Despite substantial advances in specialized deep learning hardware, data loading continues to be a major bottleneck that limits training and inference speed. To address this challenge, we propose training models directly on encoded JPEG features, reducing the computational overhead associated with full JPEG decoding and significantly improving data loading efficiency. While prior works have focused on recognition tasks, we investigate the effectiveness of this approach for the restoration task of single-image super-resolution (SISR). We present a lightweight super-resolution pipeline that operates on JPEG discrete cosine transform (DCT) coefficients in the frequency domain. Our pipeline achieves a 2.6x speedup in data loading and a 2.5x speedup in training, while preserving visual quality comparable to standard SISR approaches.