English

Weakly supervised training of universal visual concepts for multi-domain semantic segmentation

Computer Vision and Pattern Recognition 2024-03-13 v3

Abstract

Deep supervised models have an unprecedented capacity to absorb large quantities of training data. Hence, training on multiple datasets becomes a method of choice towards strong generalization in usual scenes and graceful performance degradation in edge cases. Unfortunately, different datasets often have incompatible labels. For instance, the Cityscapes road class subsumes all driving surfaces, while Vistas defines separate classes for road markings, manholes etc. Furthermore, many datasets have overlapping labels. For instance, pickups are labeled as trucks in VIPER, cars in Vistas, and vans in ADE20k. We address this challenge by considering labels as unions of universal visual concepts. This allows seamless and principled learning on multi-domain dataset collections without requiring any relabeling effort. Our method achieves competitive within-dataset and cross-dataset generalization, as well as ability to learn visual concepts which are not separately labeled in any of the training datasets. Experiments reveal competitive or state-of-the-art performance on two multi-domain dataset collections and on the WildDash 2 benchmark.

Keywords

Cite

@article{arxiv.2212.10340,
  title  = {Weakly supervised training of universal visual concepts for multi-domain semantic segmentation},
  author = {Petra Bevandić and Marin Oršić and Ivan Grubišić and Josip Šarić and Siniša Šegvić},
  journal= {arXiv preprint arXiv:2212.10340},
  year   = {2024}
}

Comments

27 pages, 16 figures, 10 tables, accepted to International Journal of Computer Vision

R2 v1 2026-06-28T07:44:49.581Z