提出分解特征统一解决图像超分辨与压缩，性能显著提升。
FIPER: Factorized Features for Robust Image Super-Resolution and Compression
- 用基-系数分解捕捉多尺度视觉特征，替代传统特征图。
- 超分辨任务平均PSNR提升204.4%，压缩任务BD-rate降低9.35%。
- 适合需要高效通用低层视觉表示的研究者与工程师。
本文提出一种统一的表示方法——分解特征（Factorized Features），用于低层视觉任务，重点测试单图像超分辨（SISR）和图像压缩。受两项任务共享恢复与保留细节原则的启发，我们采用基-系数分解及显式频率建模，以捕捉图像的结构成分与多尺度特征，解决核心挑战。将先前模型的特征表示替换为分拆特征，验证其广泛泛化潜力。此外，利用分拆特征的可合并性优化压缩流程，在多帧压缩中整合共享结构。大量实验表明，该统一表示在超分辨任务上相较基线平均提升204.4%的PSNR，图像压缩任务中相比前人最优方法降低9.35%的BD-rate。
原文摘要 · Abstract (English)
In this work, we propose using a unified representation, termed Factorized Features, for low-level vision tasks, where we test on Single Image Super-Resolution (SISR) and \textbf{Image Compression}. Motivated by the shared principles between these tasks, they require recovering and preserving fine image details, whether by enhancing resolution for SISR or reconstructing compressed data for Image Compression. Unlike previous methods that mainly focus on network architecture, our proposed approach utilizes a basis-coefficient decomposition as well as an explicit formulation of frequencies to capture structural components and multi-scale visual features in images, which addresses the core challenges of both tasks. We replace the representation of prior models from simple feature maps with Factorized Features to validate the potential for broad generalizability. In addition, we further optimize the compression pipeline by leveraging the mergeable-basis property of our Factorized Features, which consolidates shared structures on multi-frame compression. Extensive experiments show that our unified representation delivers state-of-the-art performance, achieving an average relative improvement of 204.4% in PSNR over the baseline in Super-Resolution (SR) and 9.35% BD-rate reduction in Image Compression compared to the previous SOTA. Project page: https://jayisaking.github.io/FIPER/
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。