Enhancing Quality of Compressed Images by Mitigating Enhancement Bias Towards Compression Domain
This work addresses a specific issue in image compression enhancement for applications requiring high perceptual quality, representing an incremental improvement over existing methods.
The paper tackles the problem of compressed image quality enhancement by addressing enhancement bias towards the compression domain, which degrades perceptual quality. The proposed method uses a conditional discriminator and domain-divergence regularization to distance the enhancement domain from compression, achieving superior results over state-of-the-art methods without inference overhead.
Existing quality enhancement methods for compressed images focus on aligning the enhancement domain with the raw domain to yield realistic images. However, these methods exhibit a pervasive enhancement bias towards the compression domain, inadvertently regarding it as more realistic than the raw domain. This bias makes enhanced images closely resemble their compressed counterparts, thus degrading their perceptual quality. In this paper, we propose a simple yet effective method to mitigate this bias and enhance the quality of compressed images. Our method employs a conditional discriminator with the compressed image as a key condition, and then incorporates a domain-divergence regularization to actively distance the enhancement domain from the compression domain. Through this dual strategy, our method enables the discrimination against the compression domain, and brings the enhancement domain closer to the raw domain. Comprehensive quality evaluations confirm the superiority of our method over other state-of-the-art methods without incurring inference overheads.