QualiTeacher: Quality-Conditioned Pseudo-Labeling for Real-World Image Restoration

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xiao, Fengyang, Feng, Jingjia, Hu, Peng, Zhang, Dingming, Xu, Lei, Qin, Guanyi, Li, Lu, He, Chunming, Farsiu, Sina
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908873153052672
author Xiao, Fengyang
Feng, Jingjia
Hu, Peng
Zhang, Dingming
Xu, Lei
Qin, Guanyi
Li, Lu
He, Chunming
Farsiu, Sina
author_facet Xiao, Fengyang
Feng, Jingjia
Hu, Peng
Zhang, Dingming
Xu, Lei
Qin, Guanyi
Li, Lu
He, Chunming
Farsiu, Sina
contents Real-world image restoration (RWIR) is a highly challenging task due to the absence of clean ground-truth images. Many recent methods resort to pseudo-label (PL) supervision, often within a Mean-Teacher (MT) framework. However, these methods face a critical paradox: unconditionally trusting the often imperfect, low-quality PLs forces the student model to learn undesirable artifacts, while discarding them severely limits data diversity and impairs model generalization. In this paper, we propose QualiTeacher, a novel framework that transforms pseudo-label quality from a noisy liability into a conditional supervisory signal. Instead of filtering, QualiTeacher explicitly conditions the student model on the quality of the PLs, estimated by an ensemble of complementary non-reference image quality assessment (NR-IQA) models spanning low-level distortion and semantic-level assessment. This strategy teaches the student network to learn a quality-graded restoration manifold, enabling it to understand what constitutes different quality levels. Consequently, it can not only avoid mimicking artifacts from low-quality labels but also extrapolate to generate results of higher quality than the teacher itself. To ensure the robustness and accuracy of this quality-driven learning, we further enhance the process with a multi-augmentation scheme to diversify the PL quality spectrum, a score-based preference optimization strategy inspired by Direct Preference Optimization (DPO) to enforce a monotonically ordered quality separation, and a cropped consistency loss to prevent adversarial over-optimization (reward hacking) of the IQA models. Experiments on standard RWIR benchmarks demonstrate that QualiTeacher can serve as a plug-and-play strategy to improve the quality of the existing pseudo-labeling framework, establishing a new paradigm for learning from imperfect supervision. Code will be released.
format Preprint
id arxiv_https___arxiv_org_abs_2603_08030
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle QualiTeacher: Quality-Conditioned Pseudo-Labeling for Real-World Image Restoration
Xiao, Fengyang
Feng, Jingjia
Hu, Peng
Zhang, Dingming
Xu, Lei
Qin, Guanyi
Li, Lu
He, Chunming
Farsiu, Sina
Computer Vision and Pattern Recognition
Real-world image restoration (RWIR) is a highly challenging task due to the absence of clean ground-truth images. Many recent methods resort to pseudo-label (PL) supervision, often within a Mean-Teacher (MT) framework. However, these methods face a critical paradox: unconditionally trusting the often imperfect, low-quality PLs forces the student model to learn undesirable artifacts, while discarding them severely limits data diversity and impairs model generalization. In this paper, we propose QualiTeacher, a novel framework that transforms pseudo-label quality from a noisy liability into a conditional supervisory signal. Instead of filtering, QualiTeacher explicitly conditions the student model on the quality of the PLs, estimated by an ensemble of complementary non-reference image quality assessment (NR-IQA) models spanning low-level distortion and semantic-level assessment. This strategy teaches the student network to learn a quality-graded restoration manifold, enabling it to understand what constitutes different quality levels. Consequently, it can not only avoid mimicking artifacts from low-quality labels but also extrapolate to generate results of higher quality than the teacher itself. To ensure the robustness and accuracy of this quality-driven learning, we further enhance the process with a multi-augmentation scheme to diversify the PL quality spectrum, a score-based preference optimization strategy inspired by Direct Preference Optimization (DPO) to enforce a monotonically ordered quality separation, and a cropped consistency loss to prevent adversarial over-optimization (reward hacking) of the IQA models. Experiments on standard RWIR benchmarks demonstrate that QualiTeacher can serve as a plug-and-play strategy to improve the quality of the existing pseudo-labeling framework, establishing a new paradigm for learning from imperfect supervision. Code will be released.
title QualiTeacher: Quality-Conditioned Pseudo-Labeling for Real-World Image Restoration
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2603.08030