Semantic Segmentation in Learned Compressed Domain
This work addresses efficiency issues in machine vision tasks for applications using compressed images, though it is incremental as it builds on existing compressed domain approaches.
The paper tackles the problem of suboptimal semantic segmentation performance due to distortion from image compression by proposing a compressed domain method, achieving up to 15.8% bitrate savings over compressed domain baselines and up to 83.6% bitrate and 44.8% inference time savings over pixel domain methods.
Most machine vision tasks (e.g., semantic segmentation) are based on images encoded and decoded by image compression algorithms (e.g., JPEG). However, these decoded images in the pixel domain introduce distortion, and they are optimized for human perception, making the performance of machine vision tasks suboptimal. In this paper, we propose a method based on the compressed domain to improve segmentation tasks. i) A dynamic and a static channel selection method are proposed to reduce the redundancy of compressed representations that are obtained by encoding. ii) Two different transform modules are explored and analyzed to help the compressed representation be transformed as the features in the segmentation network. The experimental results show that we can save up to 15.8\% bitrates compared with a state-of-the-art compressed domain-based work while saving up to about 83.6\% bitrates and 44.8\% inference time compared with the pixel domain-based method.