Pseudo-Label Guided Real-World Image De-weathering: A Learning Framework with Imperfect Supervision

2025-04-14

AI Summary

Real-world image deraining, dehazing, and desnowing suffer from geometric distortion artifacts due to misalignment in illumination, object positions, and scene details within real-world training pairs. To address this, we propose a pseudo-label-guided deraining framework featuring three core innovations: (1) a graph-model-driven Cross-frame Similarity Aggregation (CSA) module that explicitly models inter-frame structural consistency; (2) an Adaptive Information Allocation Strategy (IAS) that dynamically fuses ground-truth supervision with high-quality pseudo-labels; and (3) an end-to-end trainable De-Weathering network (De-W). Evaluated on non-aligned real-world rainy, snowy, and hazy datasets, our method significantly suppresses geometric distortions while enhancing texture sharpness and structural fidelity. Quantitative and qualitative results demonstrate consistent superiority over state-of-the-art methods across all weather degradation types.

Technology Category

• Computer Vision: Remote Sensing / Geospatial AI
• Machine Learning: Multi-class/Multi-label Learning & Extreme Classification
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Real-world image de-weathering aims at removingvarious undesirable weather-related artifacts, e.g., rain, snow,and fog. To this end, acquiring ideal training pairs is crucial.Existing real-world datasets are typically constructed paired databy extracting clean and degraded images from live streamsof landscape scene on the Internet. Despite the use of strictfiltering mechanisms during collection, training pairs inevitablyencounter inconsistency in terms of lighting, object position, scenedetails, etc, making de-weathering models possibly suffer fromdeformation artifacts under non-ideal supervision. In this work,we propose a unified solution for real-world image de-weatheringwith non-ideal supervision, i.e., a pseudo-label guided learningframework, to address various inconsistencies within the realworld paired dataset. Generally, it consists of a de-weatheringmodel (De-W) and a Consistent Label Constructor (CLC), bywhich restoration result can be adaptively supervised by originalground-truth image to recover sharp textures while maintainingconsistency with the degraded inputs in non-weather contentthrough the supervision of pseudo-labels. Particularly, a Crossframe Similarity Aggregation (CSA) module is deployed withinCLC to enhance the quality of pseudo-labels by exploring thepotential complementary information of multi-frames throughgraph model. Moreover, we introduce an Information AllocationStrategy (IAS) to integrate the original ground-truth imagesand pseudo-labels, thereby facilitating the joint supervision forthe training of de-weathering model. Extensive experimentsdemonstrate that our method exhibits significant advantageswhen trained on imperfectly aligned de-weathering datasets incomparison with other approaches.

Problem

• Addressing inconsistencies in real-world de-weathering datasets
• Removing weather artifacts with imperfect supervision
• Enhancing pseudo-label quality for adaptive image restoration

Innovation

• Pseudo-label guided learning framework
• Cross-frame similarity aggregation module
• Information allocation strategy integration