Papers
5
Total Citations
35
H-Index
4
About
Haowen Wang is pushing the boundaries of robotic manipulation in unstructured environments, with a research focus on 6-DoF and 7-DoF grasp pose generation, category-level pose estimation, and 3D shape reconstruction. His most influential work, "GraNet: A Multi-Level Graph Network for 6-DoF Grasp Pose Generation in Cluttered Scenes" (2023, 15 citations), introduces a graph-based approach that overcomes the limitations of non-optimized sampling methods by learning task-specific spatial features, enabling robust, object-agnostic grasping. Wang further advances the field with "DTF-Net" (2023, 6 citations), which tackles the challenge of estimating 6D poses and reconstructing shapes from RGB-D data by using deformable template fields to handle shape variations across object categories. His "Customizable 6 Degrees of Freedom Grasping Dataset" (2024, 8 citations) provides a valuable resource for training graph convolutional networks, while his work on "Self-supervised Multi-task Modeling" (2024, 2 citations) explores reconstructing articulated objects without annotated datasets. With a growing citation record and a focus on practical, model-free solutions, Wang is establishing himself as a rising figure in robotic perception and manipulation.
Research Focus
Key Achievements
Top Papers
- 1
- 2
- 3
- 4
- 5