The authors develop BeyondSCe, a zero-shot system that combines video reasoning with active view selection for event-referential grasping.
Active view selection enables event-referential grasping of occluded targets.
Zero-shot system uses event history to identify targets and selects camera viewpoints to reveal and grasp them.
Big Tech
Hyunjoon Lee · Haebeom Jung · Eunsung Cha · Daeun Lee · Yu-Chiang Frank Wang · Jaesung Choe · +1 more
Seoul National University · NVIDIA
Research Digest··2 min read
The authors present BeyondSCe, a zero-shot robotic grasping system that can resolve grasp targets defined by their role in past interactions.
Why this paper
From NVIDIA and Seoul National University
In one line
A zero-shot grasping system uses event history to identify and grasp objects referenced by past interactions, even when occluded.
What we could check
- ·No code link found
- ·No weights link found
- ·No dataset link found
- ·No compute details found
- ✓Limitations stated by the authors
- ✓Reports numbers on named benchmarks (2 benchmarks)
Observed from the paper text and links we have. Absence here means we did not find it, not that it does not exist.
§