The inaugural computer vision conference in 1990 was held over four days in Antibes, France. It had about 40 attendees, 63 accepted papers, 60 oral presentations, 20 posters, and 13 sessions.
Key points
Attendees: ~40
Papers: 63 accepted, 60 oral, 20 posters, 13 sessions
Sources
RoboWorld Challenge 2026 Launched
RoboWorld Challenge 2026 has been launched. The competition evaluates how agents ground language in visual observations, anticipate future states, and act safely. It is associated with the RoboPAD Workshop at #NeurIPS2026. Researchers and practitioners are invited to join.
Key points
Launch: RoboWorld Challenge 2026 announced.
Scope: Evaluates language grounding, future state anticipation, and safe action.
Sources
- Original post
- Linked resource - Linked in the post
Google adds Code Comprehension interview round with Gemini
Google announced a new interview round called Code Comprehension. Candidates will work on a 200‑500‑line codebase. Gemini is provided for use during the round.
Key points
New round name: Code Comprehension.
Codebase size: 200‑500 lines, with Gemini available.
Sources
Whole-Body UMI – robot‑free task skill data linked to motion prior
Whole-Body UMI collects task skills robot‑free data with UMI. It then connects them to a reusable mocap‑trained motion prior for humanoid whole‑body manipulation. No task‑specific whole‑body demos are required.
Key points
Collect task skills robot‑free data with UMI.
Connect data to a reusable mocap‑trained motion prior for humanoid whole‑body manipulation.
Sources
- Original post
- Linked resource - Linked in the post
Google AI Notetaker App for Mac Launch
Google released a free AI notetaker app for Mac. Every premium feature works without any subscription. The app is fully local and powered by Embedding Gemma 2.
Key points
Release: free AI notetaker app for Mac.
Model: powered by Embedding Gemma 2.
