Conv-Former: A Novel Network Combining Convolution and Self-Attention for Image Quality Assessment

Abstract

To address the challenge of no-reference image quality assessment (NR-IQA) for authentically and synthetically distorted images, we propose a novel network called the Combining Convolution and Self-Attention for Image Quality Assessment network (Conv-Former). Our model uses a multi-stage transformer architecture similar to that of ResNet-50 to represent appropriate perceptual mechanisms in image quality assessment (IQA) to build an accurate IQA model. We employ adaptive learnable position embedding to handle images with arbitrary resolution. We propose a new transformer block (TB) by taking advantage of transformers to capture long-range dependencies, and of local information perception (LIP) to model local features for enhanced representation learning. The module increases the model's understanding of the image content. Dual path pooling (DPP) is used to keep more contextual image quality information in feature downsampling. Experimental results verify that Conv-Former not only outperforms the state-of-the-art methods on authentic image databases, but also achieves competing performances on synthetic image databases which demonstrate the strong fitting performance and generalization capability of our proposed model.

The Image Definition Assessment of Optoelectronic Tracking Equipment Based on the BRISQUE Algorithm with Gaussian Weights

“…Comparisons are performed from two aspects: the theory and the performance of the evaluation algorithm. The main referenceless image quality evaluation algorithms that perform well are as follows: (1) Moorthy’s blind image quality index (BIQI) algorithm, which is implemented in the wavelet domain [ 25 ]; (2) Moorthy’s distortion-identification-based image verity and integrity evaluation (DIIVINE) algorithm, which is based on the BIQI algorithm [ 7 ]; (3) Saad’s distortion-identification-based image verity and integrity evaluation (DIIVINE) algorithm [ 26 ] and the BLIINDS-II improved algorithm [ 27 ]; (4) Mittal’s BRISQE algorithm [ 28 ] and the natural image quality evaluator (NIQE) algorithm, which is referenceless [ 29 ]; (5) Li’s general regression neural network (GRNN) algorithm [ 30 ]; and (6) Lintao Han’s combining convolution and self-attention for image quality assessment network [ 31 ].…”