TaskIR: Task-Driven Image Restoration via Degradation Adaptation and Task Feedback

2026-09-25

AI Summary

This study addresses the challenges of diverse degradations in real-world images and residual artifacts that impair downstream task performance by proposing a two-stage unified image restoration framework. The method introduces a degradation representation module that dynamically modulates features, integrated with a degradation-guided Transformer to achieve adaptive restoration. Furthermore, a task feedback generation and selective refinement mechanism is designed to leverage downstream task signals for eliminating residual artifacts. Experimental results demonstrate that the proposed framework simultaneously achieves competitive image restoration quality and downstream task performance across various complex degradation scenarios.

Abstract

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.

Problem

• Task-driven image restoration
• Diverse degradations
• Downstream task performance
• Unified image restoration

Innovation

• Task-driven Image Restoration
• Degradation Adaptation
• Task Feedback Refinement
• Degradation-Guided Transformer
• Unified Restoration Framework