Troubleshooting Blind Image Quality Models in the Wild
arXiv:2105.06747
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.
arXiv comment7 pages, 3 tables
References in corpus (8)
- Rethinking the Value of Network Pruning
- Blind Image Quality Assessment Using A Deep Bilinear Convolutional Neural Network
- Stabilizing the Lottery Ticket Hypothesis
- Snapshot Ensembles: Train 1, get M for free
- What Do Compressed Deep Neural Networks Forget?
- What is the State of Neural Network Pruning?
- MetaIQA: Deep Meta-learning for No-Reference Image Quality Assessment
- Active Fine-Tuning from gMAD Examples Improves Blind Image Quality Assessment