Wang Zeng

Papers

Total Citations

103

H-Index

About

Wang Zeng is a leading researcher in computer vision and 3D human modeling, with a primary focus on reconstructing detailed 3D human meshes from single 2D images. His most impactful work, "3D Human Mesh Regression With Dense Correspondence" (2020), has garnered 101 citations, establishing him as a key contributor to the field. In this seminal paper, Zeng addresses a critical limitation of prior methods that relied on global image features from convolutional neural networks (CNNs), which often lost fine-grained spatial details. His major contribution is the introduction of dense correspondence mapping, which aligns 2D image pixels directly to 3D mesh vertices, enabling more accurate and geometrically consistent human shape and pose estimation. This breakthrough has significant implications for augmented reality, human-robot interaction, and virtual try-on applications. By bridging the gap between 2D observations and 3D representations, Zeng’s work has inspired subsequent research in human mesh recovery and continues to influence both academic studies and practical deployments. His research demonstrates a commitment to solving fundamental challenges in 3D vision, making him a notable figure for students and researchers interested in human-centric computer vision.