English

What Holds Back Open-Vocabulary Segmentation?

Computer Vision and Pattern Recognition 2025-08-07 v1

Abstract

Standard segmentation setups are unable to deliver models that can recognize concepts outside the training taxonomy. Open-vocabulary approaches promise to close this gap through language-image pretraining on billions of image-caption pairs. Unfortunately, we observe that the promise is not delivered due to several bottlenecks that have caused the performance to plateau for almost two years. This paper proposes novel oracle components that identify and decouple these bottlenecks by taking advantage of the groundtruth information. The presented validation experiments deliver important empirical findings that provide a deeper insight into the failures of open-vocabulary models and suggest prominent approaches to unlock the future research.

Keywords

Cite

@article{arxiv.2508.04211,
  title  = {What Holds Back Open-Vocabulary Segmentation?},
  author = {Josip Šarić and Ivan Martinović and Matej Kristan and Siniša Šegvić},
  journal= {arXiv preprint arXiv:2508.04211},
  year   = {2025}
}

Comments

Accepted for publication at ICCV 25 Workshop: What is Next in Multimodal Foundation Models?

R2 v1 2026-07-01T04:36:51.610Z