Rishabh Saraf
Papers
2
Total Citations
17
H-Index
2
About
Rishabh Saraf is a researcher working at the exciting intersection of computer vision, natural language processing, and robotics. His primary focus is on enabling machines to understand and predict human actions, particularly in procedural tasks like cooking or assembly. Saraf’s most significant contribution is his pioneering work on zero-shot action anticipation, where he tackles the challenge of teaching robots to recognize and forecast activities they have never encountered during training. His highly cited 2022 paper, "Transferring Knowledge From Text to Video," introduces a novel hierarchical model that generalizes instructional knowledge from large-scale text corpora and transfers this understanding to the video domain. This approach allows a system to anticipate future procedural steps without ever having seen a video of that specific task. By bridging the gap between text-based knowledge and visual recognition, Saraf’s work has garnered over 17 citations and lays a critical foundation for more adaptable and intelligent robotic systems capable of learning from the vast repository of human-written instructions.
Research Focus
Key Achievements
Top Papers
- 1
- 2