Papers
2
Total Citations
45
H-Index
2
About
Deli Pei is a researcher specializing in computer vision and machine learning, with a focus on semantic image segmentation and multimodal feature learning. His work addresses the challenge of understanding visual scenes by developing efficient, unsupervised methods for segmenting images into meaningful regions. In his highly cited 2013 paper, Pei introduced an approach that leverages single-layer networks for unsupervised multimodal feature learning, combining RGB and depth images to improve semantic segmentation. This work, which has garnered 33 citations, demonstrates his contribution to reducing reliance on labeled data while enhancing segmentation accuracy through cross-modal information. Additionally, his 2013 study on efficient semantic image segmentation with multi-class ranking prior (12 citations) further advances the field by incorporating ranking mechanisms to refine segmentation outputs. Pei’s research is notable for its practical impact on autonomous systems and scene understanding, offering scalable solutions that bridge unsupervised learning and high-level visual recognition. His contributions continue to influence subsequent work in deep learning and multimodal analysis.
Research Focus
Key Achievements
Top Papers
- 1Unsupervised multimodal feature learning for semantic image segmentation33 citations · 2013
- 2Efficient semantic image segmentation with multi-class ranking prior12 citations · 2013