Automatic universal taxonomies for multi-domain semantic segmentation
This addresses the challenge of achieving proficiency across multiple visual domains for computer vision researchers, though it appears incremental as it builds on existing multi-dataset training efforts.
The paper tackles the problem of incompatible labels across multiple semantic segmentation datasets by automatically constructing universal taxonomies through iterative dataset integration, demonstrating competitive generalization performance in experiments on standard datasets.
Training semantic segmentation models on multiple datasets has sparked a lot of recent interest in the computer vision community. This interest has been motivated by expensive annotations and a desire to achieve proficiency across multiple visual domains. However, established datasets have mutually incompatible labels which disrupt principled inference in the wild. We address this issue by automatic construction of universal taxonomies through iterative dataset integration. Our method detects subset-superset relationships between dataset-specific labels, and supports learning of sub-class logits by treating super-classes as partial labels. We present experiments on collections of standard datasets and demonstrate competitive generalization performance with respect to previous work.