Siyang Sun

Beijing Academy of Artificial Intelligence, Chinese Academy of Sciences

Papers

Total Citations

125

H-Index

About

Siyang Sun is a researcher at the forefront of deep neural network optimization and autonomous systems. Their work primarily spans two critical domains: efficient deep learning model deployment and computer vision for unmanned aerial vehicles (UAVs). Sun’s most impactful contribution is a novel channel pruning method for deep neural network compression, which addresses the pressing challenge of deploying complex models on resource-constrained devices like mobile robots and smartphones. This work, with 67 citations, has become a key reference for researchers seeking to balance model accuracy with computational efficiency. In the field of autonomous aerial refueling, Sun developed a robust landmark detection and position measurement system using monocular vision, featuring a multitask parallel deep convolution neural network (MPDCNN) that achieves reliable drogue target detection. This paper, cited 54 times, demonstrates Sun’s ability to solve real-world engineering problems with innovative deep learning architectures. More recently, Sun has explored path planning for mobile robots under field-of-view constraints, ensuring observability to multiple feature points. With a growing citation impact and a focus on practical, deployable AI solutions, Siyang Sun’s work continues to influence both the theoretical foundations and applied aspects of computer vision and robotics.