Advancing Complex Video Object Segmentation via Tracking-Enhanced Prompt: The 1st Winner for 5th PVUW MOSE Challenge
arXiv:2604.00395v1 Announce Type: new Abstract: In the Complex Video Object Segmentation task, researchers are required to track and segment specific targets within cluttered environments, which rigorously tests a method's capability for target comprehension and environmental adaptability. Although SAM3, the current state-of-the-art solution, exhibits unparalleled segmentation performance and robustness on conventional targets, it underperforms on tiny and semantic-dominated objects. The root cause of this limitation lies in SAM3's insufficient comprehension of these specific target types. To address this issue, we propose TEP: Advancing Complex Video Object Segmentation via Tracking-Enhanced Prompts. As a training-free approach, TEP leverages external tracking models and Multimodal Large
View PDF HTML (experimental)
Abstract:In the Complex Video Object Segmentation task, researchers are required to track and segment specific targets within cluttered environments, which rigorously tests a method's capability for target comprehension and environmental adaptability. Although SAM3, the current state-of-the-art solution, exhibits unparalleled segmentation performance and robustness on conventional targets, it underperforms on tiny and semantic-dominated objects. The root cause of this limitation lies in SAM3's insufficient comprehension of these specific target types. To address this issue, we propose TEP: Advancing Complex Video Object Segmentation via Tracking-Enhanced Prompts. As a training-free approach, TEP leverages external tracking models and Multimodal Large Language Models to introduce tracking-enhanced prompts, thereby alleviating the difficulty SAM3 faces in understanding these challenging targets. Our method achieved first place (56.91%) on the test set of the PVUW Challenge 2026: Complex Video Object Segmentation Track.
Comments: 1st Place Solution for the 5th PVUW MOSE Challenge (CVPR 2026 Workshop)
Subjects:
Computer Vision and Pattern Recognition (cs.CV)
Cite as: arXiv:2604.00395 [cs.CV]
(or arXiv:2604.00395v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2604.00395
arXiv-issued DOI via DataCite (pending registration)
Submission history
From: Xusheng He [view email] [v1] Wed, 1 Apr 2026 02:23:23 UTC (18,147 KB)
Sign in to highlight and annotate this article
Conversation starters
Daily AI Digest
Get the top 5 AI stories delivered to your inbox every morning.
More about
modellanguage modeltraining
Unilever Deepens Commitment to AI Innovation through Vector Institute Collaboration
TORONTO, March 5, 2024 – Unilever and the Vector Institute today announced a new partnership to advance artificial intelligence (AI) innovation. Through this collaboration, Unilever will gain access to Vector [ ] The post Unilever Deepens Commitment to AI Innovation through Vector Institute Collaboration appeared first on Vector Institute for Artificial Intelligence .
CEO Update
TORONTO The Vector Institute, a leading force in artificial intelligence research and innovation, today announced that Tony Gaffney, President and CEO, will transition to a new role at [ ] The post CEO Update appeared first on Vector Institute for Artificial Intelligence .
Knowledge Map
Connected Articles — Knowledge Graph
This article is connected to other articles through shared AI topics and tags.
More in Models
Leveraging Large Language Models for More Efficient Systematic Reviews in Medicine and Beyond
New work from Vector Applied Scientist David Emerson highlights how generalist large language models (LLMs) may be used to automate systematic review screening. “Development of prompt templates for LLM-driven screening [ ] The post Leveraging Large Language Models for More Efficient Systematic Reviews in Medicine and Beyond appeared first on Vector Institute for Artificial Intelligence .
Discussion
Sign in to join the discussion
No comments yet — be the first to share your thoughts!