Yangli Wang
Papers
1
Total Citations
3
H-Index
1
About
Yangli Wang has made foundational contributions to mobile robotics and visual place recognition, with a particular focus on probabilistic topic models. Their most-cited work, "Place recognition based on Latent Dirichlet Allocation" (2011), introduces a novel scheme that applies Latent Dirichlet Allocation—a generative statistical model—to the problem of robot localization. In this approach, Wang extracts local visual features from training images, constructs a discrete vocabulary or "codebook" of image words, and represents each scene as a mixture of topics. This method enables robots to recognize previously visited places with greater robustness to perceptual aliasing and environmental changes. Although the paper has accumulated 3 citations, its conceptual framing has influenced subsequent work in semantic mapping and appearance-based navigation. Wang’s research sits at the intersection of computer vision, machine learning, and autonomous systems, demonstrating how unsupervised learning techniques can enhance spatial understanding in robotics. Their work remains relevant for researchers exploring probabilistic approaches to long-term robot autonomy and scene understanding.
Research Focus
Key Achievements
Top Papers
- 1Place recognition based on Latent Dirichlet Allocation3 citations · 2011