CVLGMar 27, 2023

Blind Image Quality Assessment via Vision-Language Correspondence: A Multitask Learning Perspective

arXiv:2303.14968v1381 citationsh-index: 73Has Code
Originality Incremental advance
AI Analysis

This work addresses the problem of accurately predicting human-perceived image quality without reference images for applications in image processing and computer vision, representing an incremental improvement over existing BIQA methods.

The authors tackled blind image quality assessment (BIQA) by developing a multitask learning scheme that automatically shares parameters and weights losses, using vision-language correspondence to incorporate auxiliary tasks like scene classification and distortion identification. The method outperformed state-of-the-art on multiple datasets, showing improved robustness and better alignment of quality annotations.

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. The source code is available at https://github.com/zwx8981/LIQE.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes