Yang He
Papers
1
Total Citations
2
H-Index
1
About
Yang He is a researcher specializing in computer vision and deep learning, with a particular focus on semantic segmentation and multi-modal scene understanding. His most notable work, "STD2P: RGBD Semantic Segmentation Using Spatio-Temporal Data-Driven Pooling" (2016), demonstrates his innovative approach to tackling complex scene parsing challenges. In this research, He proposed a superpixel-based multi-view convolutional neural network architecture that advances semantic image segmentation by intelligently integrating both spatial and temporal information from RGBD (color plus depth) data sources. His method distinguishes itself through its ability to leverage complementary views of the same scene, proving especially effective for indoor video environments where depth information enriches the semantic understanding of visual data. By combining spatio-temporal pooling strategies with deep convolutional frameworks, He's work addresses key limitations in traditional segmentation approaches, bridging the gap between single-frame analysis and richer multi-view representations. While currently accumulating citations within the research community, his contributions reflect a strong technical foundation in structured scene understanding, positioning him as an emerging voice in the application of neural networks to real-world visual perception tasks relevant to robotics, autonomous systems, and augmented reality.
Research Focus
Key Achievements
Top Papers
- 1