OV-COAST: Cost Aggregation with Optimal Transport for Open-Vocabulary Semantic Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Gandhamal, Aditya, Sikdar, Aniruddh, Sundaram, Suresh
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913874970673152
author Gandhamal, Aditya
Sikdar, Aniruddh
Sundaram, Suresh
author_facet Gandhamal, Aditya
Sikdar, Aniruddh
Sundaram, Suresh
contents Open-vocabulary semantic segmentation (OVSS) entails assigning semantic labels to each pixel in an image using textual descriptions, typically leveraging world models such as CLIP. To enhance out-of-domain generalization, we propose Cost Aggregation with Optimal Transport (OV-COAST) for open-vocabulary semantic segmentation. To align visual-language features within the framework of optimal transport theory, we employ cost volume to construct a cost matrix, which quantifies the distance between two distributions. Our approach adopts a two-stage optimization strategy: in the first stage, the optimal transport problem is solved using cost volume via Sinkhorn distance to obtain an alignment solution; in the second stage, this solution is used to guide the training of the CAT-Seg model. We evaluate state-of-the-art OVSS models on the MESS benchmark, where our approach notably improves the performance of the cost-aggregation model CAT-Seg with ViT-B backbone, achieving superior results, surpassing CAT-Seg by 1.72 % and SAN-B by 4.9 % mIoU. The code is available at https://github.com/adityagandhamal/OV-COAST/}{https://github.com/adityagandhamal/OV-COAST/ .
format Preprint
id arxiv_https___arxiv_org_abs_2506_03706
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle OV-COAST: Cost Aggregation with Optimal Transport for Open-Vocabulary Semantic Segmentation
Gandhamal, Aditya
Sikdar, Aniruddh
Sundaram, Suresh
Computer Vision and Pattern Recognition
Open-vocabulary semantic segmentation (OVSS) entails assigning semantic labels to each pixel in an image using textual descriptions, typically leveraging world models such as CLIP. To enhance out-of-domain generalization, we propose Cost Aggregation with Optimal Transport (OV-COAST) for open-vocabulary semantic segmentation. To align visual-language features within the framework of optimal transport theory, we employ cost volume to construct a cost matrix, which quantifies the distance between two distributions. Our approach adopts a two-stage optimization strategy: in the first stage, the optimal transport problem is solved using cost volume via Sinkhorn distance to obtain an alignment solution; in the second stage, this solution is used to guide the training of the CAT-Seg model. We evaluate state-of-the-art OVSS models on the MESS benchmark, where our approach notably improves the performance of the cost-aggregation model CAT-Seg with ViT-B backbone, achieving superior results, surpassing CAT-Seg by 1.72 % and SAN-B by 4.9 % mIoU. The code is available at https://github.com/adityagandhamal/OV-COAST/}{https://github.com/adityagandhamal/OV-COAST/ .
title OV-COAST: Cost Aggregation with Optimal Transport for Open-Vocabulary Semantic Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.03706