Renovating Names in Open-Vocabulary Segmentation Benchmarks

arXiv:2403.09593v1 Announce Type: new
Abstract: Names are essential to both human cognition and vision-language models. Open-vocabulary models utilize class names as text prompts to generalize to categories unseen during training. However, name qualities are often overlooked and lack sufficient precision in existing datasets. In this paper, we address this underexplored problem by presenting a framework for "renovating" names in open-vocabulary segmentation benchmarks (RENOVATE). Through human study, we demonstrate that the names generated by our model are more precise descriptions of …

abstract arxiv benchmarks class cognition cs.cv datasets however human language language models paper precision presenting prompts segmentation text training type vision vision-language models