FlowCut: Unsupervised Video Instance Segmentation via Temporal Mask Matching

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sari, Alp Eren, Favaro, Paolo
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916744171356160
author Sari, Alp Eren
Favaro, Paolo
author_facet Sari, Alp Eren
Favaro, Paolo
contents We propose FlowCut, a simple and capable method for unsupervised video instance segmentation consisting of a three-stage framework to construct a high-quality video dataset with pseudo labels. To our knowledge, our work is the first attempt to curate a video dataset with pseudo-labels for unsupervised video instance segmentation. In the first stage, we generate pseudo-instance masks by exploiting the affinities of features from both images and optical flows. In the second stage, we construct short video segments containing high-quality, consistent pseudo-instance masks by temporally matching them across the frames. In the third stage, we use the YouTubeVIS-2021 video dataset to extract our training instance segmentation set, and then train a video segmentation model. FlowCut achieves state-of-the-art performance on the YouTubeVIS-2019, YouTubeVIS-2021, DAVIS-2017, and DAVIS-2017 Motion benchmarks.
format Preprint
id arxiv_https___arxiv_org_abs_2505_13174
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FlowCut: Unsupervised Video Instance Segmentation via Temporal Mask Matching
Sari, Alp Eren
Favaro, Paolo
Computer Vision and Pattern Recognition
We propose FlowCut, a simple and capable method for unsupervised video instance segmentation consisting of a three-stage framework to construct a high-quality video dataset with pseudo labels. To our knowledge, our work is the first attempt to curate a video dataset with pseudo-labels for unsupervised video instance segmentation. In the first stage, we generate pseudo-instance masks by exploiting the affinities of features from both images and optical flows. In the second stage, we construct short video segments containing high-quality, consistent pseudo-instance masks by temporally matching them across the frames. In the third stage, we use the YouTubeVIS-2021 video dataset to extract our training instance segmentation set, and then train a video segmentation model. FlowCut achieves state-of-the-art performance on the YouTubeVIS-2019, YouTubeVIS-2021, DAVIS-2017, and DAVIS-2017 Motion benchmarks.
title FlowCut: Unsupervised Video Instance Segmentation via Temporal Mask Matching
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2505.13174