An Effective Solution for the CVPR 2026 8th UG2+ Challenge Track 3: Dynamic Object Segmentation in Turbulence

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Li, Hongzhen, Yu, Miao, Cao, Leilei, Pan, Youwei, Zhu, Yingfang, Zhu, Fengjie
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918533526454272
author Li, Hongzhen
Yu, Miao
Cao, Leilei
Pan, Youwei
Zhu, Yingfang
Zhu, Fengjie
author_facet Li, Hongzhen
Yu, Miao
Cao, Leilei
Pan, Youwei
Zhu, Yingfang
Zhu, Fengjie
contents In this work, we present our solution for the 8th UG2+ Challenge (CVPR 2026) Track 3: Dynamic Object Segmentation in Turbulence (DOST). Our method is built upon the strong baseline framework Segment Any Motion (SegAnyMo), which provides powerful mask generation and motion tracking capabilities. To further boost the segmentation performance under severe atmospheric distortions, we propose two key improvements. First, we employ a data-centric domain adaptation strategy. We significantly expand our training data by incorporating selected sequences from the DAVIS dataset alongside a subset of the DOST dataset, and apply simulated atmospheric fluctuation degradations to enhance the model's robustness against complex geometric distortions. Second, we introduce a spatio-temporal post-processing module. This refinement step effectively removes persistent boundary-connected false foregrounds and short-lived fragmented noise, while strictly preserving genuine small targets and maintaining original individual labels across frames. With these combined strategies, our proposed method ranks the 2st place in the challenge.
format Preprint
id arxiv_https___arxiv_org_abs_2606_00522
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle An Effective Solution for the CVPR 2026 8th UG2+ Challenge Track 3: Dynamic Object Segmentation in Turbulence
Li, Hongzhen
Yu, Miao
Cao, Leilei
Pan, Youwei
Zhu, Yingfang
Zhu, Fengjie
Computer Vision and Pattern Recognition
In this work, we present our solution for the 8th UG2+ Challenge (CVPR 2026) Track 3: Dynamic Object Segmentation in Turbulence (DOST). Our method is built upon the strong baseline framework Segment Any Motion (SegAnyMo), which provides powerful mask generation and motion tracking capabilities. To further boost the segmentation performance under severe atmospheric distortions, we propose two key improvements. First, we employ a data-centric domain adaptation strategy. We significantly expand our training data by incorporating selected sequences from the DAVIS dataset alongside a subset of the DOST dataset, and apply simulated atmospheric fluctuation degradations to enhance the model's robustness against complex geometric distortions. Second, we introduce a spatio-temporal post-processing module. This refinement step effectively removes persistent boundary-connected false foregrounds and short-lived fragmented noise, while strictly preserving genuine small targets and maintaining original individual labels across frames. With these combined strategies, our proposed method ranks the 2st place in the challenge.
title An Effective Solution for the CVPR 2026 8th UG2+ Challenge Track 3: Dynamic Object Segmentation in Turbulence
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2606.00522