Hongkun Tian
Papers
11
Total Citations
264
H-Index
10
About
Hongkun Tian is a robotics and computer vision researcher whose work centers on enabling robots to intelligently perceive and manipulate objects in complex, real-world environments. His research spans robotic grasp detection, instance segmentation, and RGB-D sensor fusion, with a particular focus on developing algorithms capable of handling unknown, weakly textured, and transparent objects that challenge conventional perception systems. Tian's most influential contribution, a 2022 problem-oriented review of data-driven robotic visual grasping detection (59 citations), has established him as a valuable synthesizer of this rapidly evolving field. His technical innovations include lightweight generative grasping architectures (46 citations), simultaneous instance segmentation and grasp detection frameworks, and background-adaptive grasping networks supported by novel cross-background datasets. Notably, his work on transparent and reflective objects — which pose unique difficulties even for state-of-the-art algorithms — reflects a commitment to solving pressing industrial and domestic robotics challenges. Across his body of work, Tian has accumulated over 255 citations, demonstrating meaningful influence within the robotics perception community. His consistent attention to real-world deployment constraints, including cluttered scenes, rotation invariance, and sensor limitations, makes his research particularly relevant for researchers and engineers advancing intelligent robotic manipulation systems.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10