[🔥Update - Oct 14] Our workshop will be held on 20 Oct, 8:00AM, 318A, Honolulu Convention Center. See our Workshop Schedule and Keynote Speakers!

[⚠️IMPORTANT - Aug 30] Starting from this year’s challenge, all submissions must include a valid team information file (team.json) in the submission package . Please check our Announcement for details.

[🔥Update - Aug 8] We are glad to annouance that a new track, MOSEv2, will be added to this year's challenge! Please check here for details.

[🔥Update - Aug 2] The Evaluation Servers for validation are now available!

[Update - Jul 27] Our distinguished list of Speakers has been released! Get ready to be inspired by their valuable insights and expertise!

[Update - Jul 25] The Challenge Timeline has released! Link to the evaluation server will be released on the challenge start date.

[Update - Jun 3] LSVOS 2025 will be using [OpenReview] to manage submissions. We are looking forward to your work and engaging discussions at the workshop! Please check the Paper Submission Timeline section for submission open and deadline dates.

[IMPORTANT] Team Information Requirement for Submission – Update from Previous Challenge

{
    "By submitting the test results, I agree with the challenge terms and policies.": true,
    "team_name": "Put_your_team_name_here",
    "members": [
        {"name": "Member A", "affiliation": "University A"},
        {"name": "Member B", "affiliation": "Company A"},
        {"name": "Member C", "affiliation": "Institute A"}
    ],
    "email": "your_email@example.edu"
}
  

The 7th LSVOS challenge will be held in conjunction with ICCV 2025 in Honolulu, Hawai'i. This year, we will continue the same setup as last year and still have two tracks: VOS and RVOS. In the Video Object Segmentation (VOS) track, we will utilize LVOS and MOSE. to study the VOS under more challenging complex environments. LVOS is designed for long-term videos, dealing with complex object motion and long-term reappearance, while MOSE focuses on complex scenes, covering aspects such as object disappearance and reappearance, inconspicuous small objects, heavy occlusions, and crowded environments. For the Referring Video Object Segmentation (RVOS) track, we will continue to use MeViS. MeViS focuses on the identification of the target object in a video based on motion-related descriptions rather than static attributes. This innovative approach subverts the foundational design principles of existing RVOS methods, compelling researchers to engage in a more in - depth exploration and reevaluation of motion modeling. In addition, we will hold a series of talks by the leading experts in video understating and embodied intelligence. In this year, the following topics will be covered:

• Semantic/panoptic segmentation for images/videos
• Video Object Segmentation in Complex Scenes
• Long-term Video Object Segmentation
• Referring Video Object Segmentation
• Video Segmentation with Motion Expressions
• Vision and Language
• Cognitive Models of Object Perception
• Real-world Understanding and embodied intelligence

[Update] LSVOS 2025 will be using [OpenReview] to manage submissions. We are looking forward to your work and engaging discussions at the workshop! Please check the Paper Submission Timeline section for submission open and deadline dates.

We invite authors to submit unpublished papers (8-page ICCV format) to our workshop, to be presented at a poster session upon acceptance. All submissions will go through a double-blind review process. All contributions must be submitted (along with supplementary materials, if any) through the paper submission portal.

Accepted papers will be published in the official ICCV Workshops proceedings and the Computer Vision Foundation (CVF) Open Access archive.

The 7th LSVOS challenge includes three tracks: Complex VOS (MOSEv2, new!), Classic VOS and RVOS.

Below are the links and task descriptions for the three tracks:

Congratulations to all winning teams! Check out the Technical Report of the Competition.

Track 1: Complex Video Object Segmentation (MOSEv2)

Track 2: Video Object Segmentation (Classic VOS)

Track 3: Referring Video Object Segmentation (RVOS)