Visually-rich document understanding and multimodal generation and editing of documents and text images are human-centered tasks by nature. Thus, the approaches developed to tackle them should take human interaction into account, other than the standard, and somewhat partial, evaluation scores. To engage human-centric evaluation, we will accept submissions of demos showcasing novel or published approaches. The workshop attendees will have the chance to exhibit demos during a dedicated track, enhancing the engagement and interaction of practitioners and researchers working on the workshop topics and related ones. We expect that motivated students and invited industry start-ups will be eager to participate in this special track.
Submissions
Please submit a two-page extended abstract formatted according to ICCV guidelines to our email address. Since the review process is not double-blind, be sure to include the names, affiliations, and email addresses of all authors. Your extended abstract should clearly describe both the content and format of your proposed contribution, as this information is essential for evaluation. Specifically, make sure to describe what your demo is about (possibly referencing published works, if any) and how people will interact with it. Demo sessions will take place in person, and at least one author must be present to present the work.
Deadline for the demo application submissions: 17 August, 2025 23:59 UTC-0
Author Notification: 05 September, 2025 23:59 UTC-0
Workshop date: October 19-20th, 2025