Pengxiang Gao

Papers

Total Citations

H-Index

About

Pengxiang Gao is a robotics researcher whose work lies at the intersection of computer vision and robotic manipulation, with a particular focus on enabling machines to interact with complex, real-world objects. His research spans three key areas: generative grasping, depth estimation, and deformable object manipulation. Gao’s most impactful contribution is his work on "Generative Robotic Grasping Using Depthwise Separable Convolution" (2021, 25 citations), which introduced an efficient neural network architecture for robotic grasping that balances computational speed with accuracy. He has also made significant strides in depth estimation with his "Two-stage deep regression enhanced depth estimation from a single RGB image" (2020, 15 citations), addressing a critical challenge for applications like autonomous driving and augmented reality where traditional depth sensors are impractical. Notably, Gao’s work on "Multidimensional Deformable Object Manipulation Based on DN-Transporter Networks" (2022, 14 citations) tackles the notoriously difficult problem of handling non-rigid objects such as cables and packaging materials—a crucial capability for logistics and manufacturing. With over 54 citations across his most-cited works, Gao is establishing himself as a rising figure in robotic perception and manipulation, bridging the gap between theoretical computer vision and practical robotic applications.