Demos and Papers

Calls

We invite submissions of extended abstracts. We encourage workshop contributions to be accompanied by a demo to be shown during the workshop. While live demos—such as web-based ones or those running on hardware like phones or laptops are highly encouraged, for this first edition of the workshop, video demos will also be accepted to accommodate any challenges participants may face in transporting their demo setups or devices to the workshop. The review process will be double-blind. Participants will be asked to upload their extended abstracts (with demo plans) to CMT.

Topics

We invite contributions on topics relevant to the development of vision-based assistants such as, (but not limited to):

  • Streaming/online vision-language models
  • Real-time activity understanding
  • Grounding of vision-language models
  • Ego-centric video understanding
  • Language and robot learning

Submission Instructions

  • Submission Link: CMT
  • Submission Types:
    • Extended Abstracts: The maximum page length is four pages, with additional pages for demo plans (if available) and references. We encourage the submission relevant work that are in progress and those that have been previously (recently) published. As subject area please choose “Paper -> Extended Abstract”.
    • Papers Accepted to CVPR 2025: Full length (8-page) papers can be submitted, with additional pages for demo plans (if available) and references. As subject area please choose “Paper -> Accepted to CVPR 2025”.
  • Paper Template: The CVPR 2025 author kit is available here.

All submissions are be non-archival. Supplementary material can be added to the same file (after demo plans and/or references).

Dates

  • Paper Submission Deadline: May 1st, 2025
  • Notification to Authors: May 15th, 2025
  • Camera-ready deadline: June 1st, 2025

All deadlines are until 11:59 pm Pacific Standard Time (PST).