Hongen Shao

Central South University of Forestry and Technology, South China University of Technology

Papers

Total Citations

H-Index

About

Hongen Shao is a researcher working at the intersection of multimodal artificial intelligence and efficient 3D computer vision, with contributions spanning emotion recognition systems and hardware-accelerated deep learning. Their most impactful work focuses on Multimodal Emotion Recognition in Conversation (MERC), where their 2024 paper on Masked Graph Learning with Recurrent Alignment has garnered 39 citations, demonstrating significant community interest in their approach to fusing complementary semantic information across modalities for applications in sentiment monitoring and intelligent dialogue systems. Shao has also made meaningful contributions to the efficiency of point cloud neural networks, a critical challenge for real-world deployment in autonomous driving, robotics, and virtual reality. Their work on exploiting geometric similarity for point cloud acceleration (14 citations) and the subsequent SimDiff framework, which leverages spatial similarity and differential execution, collectively advance the feasibility of running 3D vision models on resource-constrained edge devices. With a growing citation record across two distinct yet complementary research domains, Shao represents an emerging voice in applied AI research, bridging perceptual intelligence with the practical demands of real-time, energy-efficient computation.