Troubleshooting Blind Image Quality Models in the Wild
- URL: http://arxiv.org/abs/2105.06747v1
- Date: Fri, 14 May 2021 10:10:48 GMT
- Title: Troubleshooting Blind Image Quality Models in the Wild
- Authors: Zhihua Wang and Haotao Wang and Tianlong Chen and Zhangyang Wang and
Kede Ma
- Abstract summary: Group maximum differentiation competition (gMAD) has been used to improve blind image quality assessment (BIQA) models.
We construct a set of "self-competitors," as random ensembles of pruned versions of the target model to be improved.
Diverse failures can then be efficiently identified via self-gMAD competition.
- Score: 99.96661607178677
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: Recently, the group maximum differentiation competition (gMAD) has been used
to improve blind image quality assessment (BIQA) models, with the help of
full-reference metrics. When applying this type of approach to troubleshoot
"best-performing" BIQA models in the wild, we are faced with a practical
challenge: it is highly nontrivial to obtain stronger competing models for
efficient failure-spotting. Inspired by recent findings that difficult samples
of deep models may be exposed through network pruning, we construct a set of
"self-competitors," as random ensembles of pruned versions of the target model
to be improved. Diverse failures can then be efficiently identified via
self-gMAD competition. Next, we fine-tune both the target and its pruned
variants on the human-rated gMAD set. This allows all models to learn from
their respective failures, preparing themselves for the next round of self-gMAD
competition. Experimental results demonstrate that our method efficiently
troubleshoots BIQA models in the wild with improved generalizability.
Related papers
- LLM-Assisted Red Teaming of Diffusion Models through "Failures Are Fated, But Can Be Faded" [7.736445799116692]
"Failures are fated, but can be faded" is a framework to explore and construct the failure landscape in pre-trained generative models.
We show how to restructure the failure landscape to be more desirable by moving away from the discovered failure modes.
arXiv Detail & Related papers (2024-10-22T06:46:09Z) - Adaptive Image Quality Assessment via Teaching Large Multimodal Model to Compare [99.57567498494448]
We introduce Compare2Score, an all-around LMM-based no-reference IQA model.
During training, we generate scaled-up comparative instructions by comparing images from the same IQA dataset.
Experiments on nine IQA datasets validate that the Compare2Score effectively bridges text-defined comparative levels during training.
arXiv Detail & Related papers (2024-05-29T17:26:09Z) - Opinion-Unaware Blind Image Quality Assessment using Multi-Scale Deep Feature Statistics [54.08757792080732]
We propose integrating deep features from pre-trained visual models with a statistical analysis model to achieve opinion-unaware BIQA (OU-BIQA)
Our proposed model exhibits superior consistency with human visual perception compared to state-of-the-art BIQA models.
arXiv Detail & Related papers (2024-05-29T06:09:34Z) - Comparison of No-Reference Image Quality Models via MAP Estimation in
Diffusion Latents [99.19391983670569]
We show that NR-IQA models can be plugged into the maximum a posteriori (MAP) estimation framework for image enhancement.
Different NR-IQA models are likely to induce different enhanced images, which are ultimately subject to psychophysical testing.
This leads to a new computational method for comparing NR-IQA models within the analysis-by-synthesis framework.
arXiv Detail & Related papers (2024-03-11T03:35:41Z) - Black-box Adversarial Attacks Against Image Quality Assessment Models [16.11900427447442]
The goal of No-Reference Image Quality Assessment (NR-IQA) is to predict the perceptual quality of an image in line with its subjective evaluation.
This paper makes the first attempt to explore the black-box adversarial attacks on NR-IQA models.
arXiv Detail & Related papers (2024-02-27T14:16:39Z) - Towards Calibrated Robust Fine-Tuning of Vision-Language Models [97.19901765814431]
This work proposes a robust fine-tuning method that improves both OOD accuracy and confidence calibration simultaneously in vision language models.
We show that both OOD classification and OOD calibration errors have a shared upper bound consisting of two terms of ID data.
Based on this insight, we design a novel framework that conducts fine-tuning with a constrained multimodal contrastive loss enforcing a larger smallest singular value.
arXiv Detail & Related papers (2023-11-03T05:41:25Z) - Uncertainty-Aware Blind Image Quality Assessment in the Laboratory and
Wild [98.48284827503409]
We develop a textitunified BIQA model and an approach of training it for both synthetic and realistic distortions.
We employ the fidelity loss to optimize a deep neural network for BIQA over a large number of such image pairs.
Experiments on six IQA databases show the promise of the learned method in blindly assessing image quality in the laboratory and wild.
arXiv Detail & Related papers (2020-05-28T13:35:23Z) - Comparison of Image Quality Models for Optimization of Image Processing
Systems [41.57409136781606]
We use eleven full-reference IQA models to train deep neural networks for four low-level vision tasks.
Subjective testing on the optimized images allows us to rank the competing models in terms of their perceptual performance.
arXiv Detail & Related papers (2020-05-04T09:26:40Z) - Active Fine-Tuning from gMAD Examples Improves Blind Image Quality
Assessment [29.196117743753813]
We show that gMAD examples can be used to improve blind IQA (BIQA) methods.
Specifically, we first pre-train a DNN-based BIQA model using multiple noisy annotators.
We then seek pairs of images by comparing the baseline model with a set of full-reference IQA methods in gMAD.
arXiv Detail & Related papers (2020-03-08T21:19:01Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.