MANIQA: Multi-dimension Attention Network for No-Reference Image Quality Assessment

By Sidi Yang, Tianhe Wu, Shu Shi, S. Gong, Ming Cao, Jiahao Wang, Yujiu Yang. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2022

759 citations recorded by the source corpus.

No-Reference Image Quality Assessment (NR-IQA) aims to assess the perceptual quality of images in accordance with human subjective perception. Unfortunately, existing NR-IQA methods are far from meeting the needs of predicting accurate quality scores on GAN-based distortion images. To this end, we propose Multi-dimension Attention Network for no-reference Image Quality Assessment (MANIQA) to improve the performance on GAN-based distortion. We firstly extract features via ViT, then to strengthen global and local interactions, we propose the Transposed Attention Block (TAB) and the Scale Swin Transformer Block (SSTB). These two modules apply attention mechanisms across the channel and spatial dimension, respectively. In this multi-dimensional manner, the modules cooperatively increase the interaction among different regions of images globally and locally. Finally, a dual branch structure for patch-weighted quality prediction is applied to predict the final score depending on the weight of each patch’s score. Experimental results demonstrate that MANIQA outperforms state-of-the-art methods on four standard datasets (LIVE, TID2013, CSIQ, and KADID-10K) by a large margin. Besides, our method ranked first place in the final testing phase of the NTIRE 2022 Perceptual Image Quality Assessment Challenge Track 2: No-Reference. Codes and models are available at https://github.com/IIGROUP/MANIQA.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/d2171803c7aae785f05e9f298f6025f4252262c6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Chen Change Loy
3. Yu Qiao
4. Dacheng Tao
5. Ming-Hsuan Yang
6. Lei Zhang
7. Xiaogang Wang
8. Wanli Ouyang

In Media Technology

1. Xiaokang Yang
2. Jinjin Gu
3. Xintao Wang
4. Guangtao Zhai
5. Chao Dong
6. Yu Qiao
7. Weixia Zhang
8. Kede Ma

Citations in the corpus

• SeeSR: Towards Semantics-Aware Real-World Image Super-Resolution (2023, neutral): “NIQE [71], MANIQA [64], MUSIQ [64], and CLIPIQA [64] are no-reference image quality measures.”
• DiffBIR: Towards Blind Image Restoration with Generative Diffusion Prior (2023, methodology): “To better evaluate the realness for BIR task, we also include several no-reference image quality assessment (IQA) metrics: MANIQA4 [60] and NIQE.”
• Scaling Up to Excellence: Practicing Model Scaling for Photo-Realistic Image Restoration In the Wild (2024, methodology): “We selected the following metrics for quantitative comparison: full-reference metrics PSNR, SSIM, LPIPS [93], and the non-reference metrics ManIQA [83], ClipIQA [74], MUSIQ [43].”
• Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspective (2023, background): “Over the years, BIQA methods have explored different computational structures: generalized divisive normalization [37], multi-level feature aggregation [25], adaptive convolution [55], and self-attention [13, 19, 70]; and objective functions: p-norm induced metric, fidelity-based ranking losses [58]”
• OminiControl: Minimal and Universal Control for Diffusion Transformer (2024, positive): “Generation quality is measured using FID [16], SSIM, CLIP-IQA[53], MAN-IQA [59], MUSIQ [21], and PSNR[7] for visual fidelity, along with CLIP Text and CLIP Image scores [43] for consistency.”