Dehao Huang

Southern University of Science and Technology

Papers

8

Total Citations

169

H-Index

5

About

Dehao Huang is a leading researcher in robotic manipulation, with a primary focus on **task-oriented grasping (TOG)**—the challenge of enabling robots to grasp objects in a way that is compatible with a subsequent manipulation task. His work sits at the intersection of **computer vision, natural language processing, and robotics**, pioneering the integration of large language models and foundation models into grasping pipelines. Huang’s most impactful contribution is **GraspGPT** (2023, 88 citations), which leverages semantic knowledge from large language models to predict task-appropriate grasps, setting a new standard for TOG. He further advanced the field with **FoundationGrasp** (2025, 22 citations), demonstrating how foundation models can generalize grasping strategies across diverse objects and tasks without task-specific training. His work on **visual-language inputs for grasp prediction** (38 citations) bridges the gap between human commands and robotic action by jointly solving object grounding and task grounding. Beyond grasping, Huang has contributed to **cloud-based VSLAM** and **efficient object rearrangement via multi-view fusion**, showcasing a broader interest in autonomous navigation and scene understanding. With over 160 citations and a string of recent publications (2023–2025), Dehao Huang is at the forefront of making assistive robots that can understand language, reason about tasks, and manipulate tools intelligently in human environments.

Research Focus

Key Achievements

5
H-Index
8
Papers
169
Total Citations
21
Avg Citations/Paper
🏆 Most Cited Paper
GraspGPT: Leveraging Semantic Knowledge From a Large Language Model for Task-Oriented Grasping
88 citations · 2023
📈 Most Prolific Year: 2023 (4 Papers)
🤝 Key Collaborators: 18
🏛 Institutions: Southern University of Science and Technology

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 13 days ago