The focus of this workshop is on embodied visual tasks that require the grounding of language to actions in real-world settings. Specifically, we want to draw focus to challenges like partial observability, continuous state spaces, and irrevocable actions for language-guided agents in visual environments.
Such challenges are not captured by current datasets for grounding and embodiment [1, 2, 3].
Egocentric and Robotic vision
Navigation and Motion Planning
Learning from Demonstration
Task and Symbolic Planning
Deep Reinforcement Learning
To encourage research in embodied vision & language, the workshop includes a benchmark challenge based on ALFRED.
This benchmark captures real-world complexities like object state changes, and requires long-horizon planning. This workshop exists to bring together Vision, Robotics, and NLP researchers to tackle the unique challenges of this three-field intersection that are often avoided when focusing only on vision-and-language or vision-and-robotics (i.e., 'embodied AI').
Challenge Papers Participants are required to upload their model to our evaluation server. The evaluation server automatically evaluates the models on an unseen test set. Final numbers for the prize challenge will be frozen on Aug 5
Publication OptionsArchival vs Unofficial Papers can submitted for publication in either the official proceedings (archival) or to be hosted on this website (unofficial). Both submission types can be presented at the workshop, but opting out of the proceedings allows you to submit your work for publication at another venue. Unofficial submissions are not required to be in the ECCV format. Please add either Archival or Unofficial as a keyword when submitting to indicate the correct submission track. No submissions will be made public during the review process. If there is any doubt we will contact authors to confirm the desired submission track.
Prize Leaderboard closes
Aug 10 (5pm PDT)
Abstract Deadline ECCV conference organizers have asked that there be a version of all papers (official and unofficial) in the conference system by Aug 5. We are interpreting this restriction as abstract only and that you will be allowed to make changes via OpenReview to the full paper through the deadline and all papers can be heavily revised for the camera ready. We are sorry for the new virtual restrictions. We will continue to update this space (including how to upload abstracts) as the conference organizers give us more information. Please don't hesitate to reach out with questions!
NEW! System Descriptions Submit your descriptions to OpenReview by Aug 10 (5pm Pacific Daylight Time).
Do unofficial submissions need to be in ECCV format? No
Can unofficial submissions already be published? Yes
Page lengths for official submissions? 4 - 14 pages (ECCV rules require not major overlap with existing published work)
Page lengths for unofficial submissions? Minimum 4 pages
Is reviewing blind? Official are double blind, Unofficial are single
Supplemental Material Can be merged to the end of the submission PDF
Competition Metric Ranking and awards are based on Unseen Success Rate