The official repo of Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspective (CVPR2023)

We aim at advancing blind image quality assessment (BIQA), which predicts the human perception of image quality without any reference information. We develop a general and automated multitask learning scheme for BIQA to exploit auxiliary knowledge from other tasks, in a way that the model parameter sharing and the loss weighting are determined automatically. Specifically, we first describe all candidate label combinations (from multiple tasks) using a textual template, and compute the joint probability from the cosine similarities of the visual-textual embeddings. Predictions of each task can be inferred from the joint distribution, and optimized by carefully designed loss functions. Through comprehensive experiments on learning three tasks - BIQA, scene classification, and distortion type identification, we verify that the proposed BIQA method 1) benefits from the scene classification and distortion type identification tasks and outperforms the state-of-the-art on multiple IQA datasets, 2) is more robust in the group maximum differentiation competition, and 3) realigns the quality annotations from different IQA datasets more effectively.

torch 1.8+

torchvision

Python 3

pip install ftfy regex tqdm

pip install git+https://github.com/openai/CLIP.git

python train_unique_clip_weight.py

python BIQA_benchmark.py

python demo.py

python demo2.py

Google Drive:

https://drive.google.com/file/d/1GoKwUKNR-rvX11QbKRN8MuBZw2hXKHGh/view?usp=sharing

百度网盘：

链接: https://pan.baidu.com/s/1KHjj7T8y2H_eKE6w7HnWJA 提取码: 2b8v

IQA-PyTorch supports LIQE now! Can be easily used as follows:

import pyiqa
model = pyiqa.create_metric('liqe', as_loss=False) #Re-trained on the official set of KonIQ-10k
score = model(img_path)

or

import pyiqa
model = pyiqa.create_metric('liqe_mix', as_loss=False) #Trained on multiple datasets as in the paper.
score = model(img_path)

python train_liqe_single.py # it will train on official split of KonIQ-10k for fair comparison with other work

@inproceedings{zhang2023liqe,  
  title={Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspective},  
  author={Zhang, Weixia and Zhai, Guangtao and Wei, Ying and Yang, Xiaokang and Ma, Kede},  
  booktitle={IEEE Conference on Computer Vision and Pattern Recognition},  
  pages={14071--14081},
  year={2023}
}