[visionlist] Commands for Autonomous Vehicles Workshop @ ECCV2020

[Simon Vandenhende]

Dear Researchers,


We are happy to invite you to join the first 'Commands for Autonomous Vehicles Workshop' (C4AV). The topic of the workshop is multi-modal learning - language and vision - in a practical setting. Participants are invited to share their work as a poster or oral presentation during the workshop (see call for papers -  https://c4av-2020.github.io/).


We will be accepting papers in the following areas:

  *   Visual Dialog

  *   Multi-modal feature learning

  *   Object Referral/Visual Grounding

  *   Visual Question Answering

  *   Embodied Question Answering

  *   Zero-shot/Few-shot in multi-modal learning

  *   Applications in joint text/image understanding

  *   ...

Additionally, we want to invite you to compete in the C4AV challenge. Given a natural language command that expresses an action that needs to be taken by an autonomous vehicle, participants need to develop a model that identifies the referred object in the scene. The challenge is based on the Talk2Car dataset (EMNLP 2019), which extends the nuScenes dataset for autonomous driving with a visual grounding task. Top performing teams can win prizes and will be invited to present their work at the C4AV workshop at ECCV2020.


For more information visit https://c4av-2020.github.io/ . We are looking forward to seeing you on the leaderboard.


The C4AV team

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://visionscience.com/pipermail/visionlist_visionscience.com/attachments/20200415/f7e9b1a0/attachment.html>