Abstract
Evaluating image quality without reference images, known as blind image quality assessment (BIQA), is crucial for image communication. Recently, convolutional neural networks (CNNs) have emerged as a prominent BIQA approach due to their feature learning power. Usually, both high-level semantic information and low-level details significantly impact perceived visual quality. However, most existing CNN-based methods focus on high-level semantic information via aggregating features on top of the last convolutional layer into a global descriptor, neglecting the importance of shallow, low-level cues. To address this limitation, this paper proposes a novel approach that exploits local encoding and histogram-based pyramid pooling on crosslayer features produced by a CNN, achieving a joint local and global analysis. Specifically, we introduce a cross-layer pattern encoding model that characterizes features generated along convolutional layers via a soft histogram of local 3D binary patterns. This leads to a highly informative yet compact descriptor for score regression. By building this module into a ResNet backbone, we present an effective BIQA model demonstrating state-ofthe-art performance in extensive experiments on synthetic and authentic datasets.
Funding Information
  • National Natural Science Foundation of China (62401210)
  • Natural Science Foundation of Guangdong Province (2025A1515011539)
  • Basic and Applied Basic Research Foundation of Guangdong Province (2023A1515110646)
  • Guangzhou Basic and Applied Basic Research Project (2025A04J4502)
  • National Key Research and Development Program of China (2024YFE0105400)
  • National Natural Science Foundation of China (62472179)
  • Guangzhou Science and Technology Plan Project - Key R and D Plan (2024B01W0007)
  • Key Research and Development Project of Guangzhou (202206010091)
  • Fund of Southern Marine Science and Engineering Guangdong Laboratory at Zhanjiang (ZJW-2023-04)
  • Special Fund for Marine Economic Development of Guangdong Province, China (GDNRC[2024])