Ming–Hsuan Yang

About

Ming-Hsuan Yang is a leading researcher in computer vision, with a primary focus on 3D object pose estimation, visual tracking, and simultaneous localization and mapping (SLAM). His most influential work, “Multiple View Feature Descriptors from Image Sequences via Kernel Principal Component Analysis” (2004), has garnered 40 citations and introduced a novel method for extracting robust feature descriptors from multiple views, significantly improving wide-baseline matching under varying illumination and viewpoint conditions. Yang’s contributions to 6DoF object pose tracking are particularly notable; his benchmark dataset (2017, 29 citations) provides a critical standard for evaluating real-time pose tracking algorithms in augmented reality and robotics. He further advanced direct pose estimation techniques for planar objects (2018, 10 citations; 2016, 6 citations), addressing the limitations of traditional Perspective-n-Point methods by eliminating the need for explicit feature extraction. His work on vision-based SLAM (2005, 16 citations) integrates multiple-view descriptors to enhance mapping robustness. More recently, Yang has explored dual robotic arm path planning (2023), demonstrating the breadth of his expertise from theoretical computer vision to practical robotics applications.

Research Focus

Key Achievements

Top Papers

1. 1
   Multiple View Feature Descriptors from Image Sequences via Kernel Principal Component Analysis

   40 citations · 2004
2. 2
   [POSTER] A Benchmark Dataset for 6DoF Object Pose Tracking

   29 citations · 2017
3. 3
   Simultaneous localization and mapping using multiple view feature descriptors

   16 citations · 2005
4. 4
   Direct pose estimation for planar objects

   10 citations · 2018
5. 5
   Direct 3D pose estimation of a planar target

   6 citations · 2016
6. 6
   Research on Dual Robotic Arm Path Planning Based on Steering Wheel Sewing Device

   3 citations · 2023

Key Collaborators

Contact & Links