发表机构
Northwestern Polytechnical University; Shenzhen Research Institute of Northwestern Polytechnical University（西北工业大学; 西北工业大学深圳研究院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
TaskIR提出两阶段任务驱动图像恢复框架，结合退化自适应与任务反馈细化，提升多样退化下的恢复质量和下游任务性能。
AI 中文摘要
任务驱动的图像恢复旨在同时提升图像质量和下游任务性能。然而，现有方法主要集中于单一退化类型，难以应对现实场景中遇到的各种退化。不同的退化对恢复提出了不同的要求，恢复不足可能会留下残余退化和伪影，从而损害物体边界和语义线索，进而影响下游任务性能。为解决这些挑战，我们提出了TaskIR，一个两阶段的任务驱动统一图像恢复框架，该框架将退化自适应恢复与任务反馈细化相结合。在第一阶段，退化表示模块（DRM）提取退化表示，使退化引导的Transformer块（DGTB）能够动态调整特征变换以实现自适应恢复。在第二阶段，任务到恢复反馈生成模块（TRFG）通过建模与当前恢复相关的任务表示差异，将异构任务特征转换为恢复反馈。随后，选择性任务反馈细化模块（STFR）评估反馈相关性，并选择性地细化中间恢复特征，以减轻对已良好恢复内容的干扰。大量实验表明，TaskIR在各种退化和任务中均取得了具有竞争力的恢复质量和下游任务性能。
英文摘要
Task-driven image restoration aims to improve both image quality and downstream task performance. However, existing methods predominantly focus on single degradation type and struggle to handle the diverse degradations encountered in real-world scenarios. Different degradations impose distinct restoration demands, and insufficient restoration may leave residual degradations and artifacts that impair object boundaries and semantic cues, thereby compromising downstream task performance. To address these challenges, we propose TaskIR, a two-stage task-driven unified image restoration framework that integrates degradation-adaptive restoration with task feedback refinement. In Stage I, a Degradation Representation Module (DRM) extracts degradation representations, enabling a Degradation-Guided Transformer Block (DGTB) to dynamically modulate feature transformations for adaptive restoration. In Stage II, a Task-to-Restoration Feedback Generation module (TRFG) transforms heterogeneous task features into restoration feedback by modeling task-representation discrepancies associated with the current restoration. Subsequently, a Selective Task Feedback Refinement module (STFR) assesses feedback relevance and selectively refines intermediate restoration features to mitigate interference with well-restored content. Extensive experiments demonstrate that TaskIR achieves competitive restoration quality and downstream task performance across diverse degradations and tasks.
