Jinyu He
Papers
1
Total Citations
1
H-Index
1
About
Jinyu He is a rising researcher at the forefront of 3D computer vision and robotics perception, with a focus on bridging the gap between 2D foundation models and 3D scene understanding. His most notable work, "RE0: Recognize Everything with 3D Zero-Shot Instance Segmentation" (2025), tackles one of the field’s hardest challenges: enabling robots to recognize and segment arbitrary objects in 3D without task-specific training. By leveraging vision foundation models (VFMs) to overcome the scarcity of high-quality 3D data, He proposes a novel framework that transfers 2D recognition capabilities into the 3D domain, achieving zero-shot generalization. While his citation count is still growing, this work has already attracted attention for its practical implications in open-world robotics. He’s contributions are particularly significant for advancing embodied AI, where agents must interact with unseen objects in unstructured environments. His research sits at the intersection of zero-shot learning, instance segmentation, and 3D representation, promising to make robots more adaptable and intelligent in real-world settings.
Research Focus
Key Achievements
Top Papers
- 1RE0: Recognize Everything with 3D Zero-Shot Instance Segmentation1 citations · 2025