Sicheng Zhao
Papers
3
Total Citations
87
H-Index
2
About
Sicheng Zhao is a prominent researcher working at the intersection of affective computing, embodied artificial intelligence, and multimodal perception. His work spans some of the most dynamic frontiers in modern AI, with particular focus on enabling machines to understand human emotions, interact intelligently with physical environments, and interpret three-dimensional space from visual inputs. Zhao's most influential contribution, "Unlocking the Emotional World of Visual Media" (2023, 77 citations), offers a comprehensive overview of artificial emotional intelligence, examining how deep learning is transforming computers' and robots' capacity to recognize and respond to human affect — a breakthrough with profound implications for human-computer interaction. His survey on embodied learning for object-centric robotic manipulation (2025) reflects his growing engagement with next-generation intelligent robotics, addressing how agents can learn through physical interaction rather than purely data-driven approaches. More recently, his work on LLMI3D explores how multimodal large language models can achieve robust 3D perception from single 2D images, tackling generalization challenges critical for autonomous driving and augmented reality applications. Together, these contributions position Zhao as a versatile and forward-thinking researcher whose work bridges perception, emotion, and embodied intelligence — making him an essential voice in contemporary AI research.
Research Focus
Key Achievements
Top Papers
- 1
- 2A Survey of Embodied Learning for Object-centric Robotic Manipulation8 citations · 2025
- 3LLMI3D: MLLM-based 3D Perception from a Single 2D Image2 citations · 2024