Papers
7
Total Citations
538
H-Index
7
About
Xinke Deng is a leading researcher in robot perception and manipulation, whose work has fundamentally advanced how robots understand and interact with their physical environment. Her primary research areas include 6D object pose estimation and tracking, self-supervised learning for robotics, and visual simultaneous localization and mapping (SLAM). Deng’s most significant contribution is the development of PoseRBPF, a Rao-Blackwellized particle filter framework for 6D object pose tracking that has garnered over 268 citations across its publications. This work provides a principled probabilistic approach to tracking object rotations and translations from video, enabling robots to perform complex manipulation and navigation tasks with unprecedented accuracy. Her highly cited paper on self-supervised 6D object pose estimation (198 citations) introduced a groundbreaking robot system that learns from unlabeled real-world data, dramatically reducing the need for expensive manual annotation. Deng has also made notable contributions to active visual SLAM for mobile robot navigation and goal-based imitation learning, where she developed methods for robots to infer symbolic goals from third-person video demonstrations. Her research bridges the gap between theoretical probabilistic methods and practical robotic applications, making her work essential reading for anyone interested in robot manipulation, computer vision, or autonomous systems.
Research Focus
Key Achievements
Top Papers
- 1Self-supervised 6D Object Pose Estimation for Robot Manipulation198 citations · 2020
- 2PoseRBPF: A Rao–Blackwellized Particle Filter for 6-D Object Pose Tracking166 citations · 2021
- 3PoseRBPF: A Rao-Blackwellized Particle Filter for 6D Object Pose Tracking52 citations · 2019
- 4PoseRBPF: A Rao-Blackwellized Particle Filter for6D Object Pose Estimation50 citations · 2019
- 5Feature-constrained Active Visual SLAM for Mobile Robot Navigation46 citations · 2018
- 6Self-supervised 6D Object Pose Estimation for Robot Manipulation14 citations · 2019
- 7Motion Reasoning for Goal-Based Imitation Learning12 citations · 2020