Search papers, labs, and topics across Lattice.
This paper introduces a new task called Anomaly Perception for UGC image Enhancement (UEAP) and presents the UEAP-4k benchmark dataset, which includes fine-grained annotations for various anomaly types in enhanced images. The authors propose the Difference-Fusion Anomaly Perception Method (DFAP-UGC), which utilizes a novel approach combining problem-reference difference fusion with dynamic task prioritization to effectively identify localized anomalies in UGC-enhanced images. Experimental results demonstrate that DFAP-UGC significantly outperforms existing methods, highlighting the importance of addressing specific visual anomalies to maintain viewer trust in enhanced content.
UGC-enhanced images can harbor subtle anomalies that existing quality assessment methods completely overlook, but our new framework identifies these issues with unprecedented precision.
Image enhancement and restoration have become standard back-end operations on short-video and social media platforms to boost UGC visual experience. Yet these processes inevitably introduce visual anomalies--especially in faces, texts, and textures--that directly undermine perceptual fidelity and viewer trust. While existing IQA methods perform well on classic distortions, they target holistic quality assessment and fail to capture the specific, localized anomalies caused by enhancement algorithms in real-world UGC. To bridge this gap, we formally define a new task-quality Anomaly Perception for UGC image Enhancement (UEAP), and contribute the first UEAP benchmark dataset, named UEAP-4k, curated from the real business scenarios. It provides fine-grained annotations for anomaly categories, localization and severity levels. Furthermore, we propose a Difference-Fusion Anomaly Perception Method (DFAP-UGC) for wild UGC-enhanced images, which leverages explicit problem-reference difference fusion with dense spatial querying, regional verification, and quality-aware ranking, enabling robust anomaly identification in challenging scenarios. To handle the inherent coupling of subtasks in this new task, we propose a Locality-Aware Dynamic Task Prioritization (LADTP) training strategy that enables effective end-to-end learning and eliminates multi-stage overhead. Extensive experiments show that our method outperforms baselines adapted from classical approaches for this task, validating the value of this dataset and the superior of DFAP-UGC for robust UGC-enhanced image anomaly perception. Code and data will be public.
