Papers
2
Total Citations
34
H-Index
2
About
Yan Bin Ng’s research focuses on the cutting-edge intersection of computer vision and human activity understanding, with a particular emphasis on **future action forecasting** from video data. His major contribution lies in developing novel approaches to predict sequences of actions that will complete an ongoing activity, even when only partial observations are available. In his most-cited work, “Forecasting Future Action Sequences With Attention: A New Approach to Weakly Supervised Action Forecasting” (2020, 32 citations), Ng introduced an attention-based neural machine translation framework that significantly advances the ability to anticipate unseen future actions—a critical capability for applications in assistive robotics, video surveillance, and security. His earlier foundational paper (2019, 2 citations) laid the groundwork for this approach, demonstrating how to forecast complete action sequences from partial video inputs. By addressing the challenging problem of weakly supervised learning, Ng’s work enables more practical and scalable systems that can learn from limited labeled data. His research has direct implications for creating smarter, more responsive AI systems that can proactively assist humans in real-world environments, marking him as an emerging voice in the field of predictive visual intelligence.
Research Focus
Key Achievements
Top Papers
- 1
- 2Forecasting Future Sequence of Actions to Complete an Activity.2 citations · 2019