Hui Miao

Papers

Total Citations

H-Index

About

Hui Miao is a researcher working at the intersection of computer vision, robotics, and fine-grained visual understanding. Their most notable work, "3D Part Guided Image Editing for Fine-Grained Object Understanding" (2020), addresses a fundamental challenge in enabling robots and autonomous systems to perceive and interact meaningfully with the physical world. By developing methods that leverage 3D part structures to guide image editing, Miao's research advances how visual models can decompose objects into their constituent movable components — a capability critical for applications like self-driving vehicles, where understanding the dynamic behavior of surrounding objects (such as opening car doors or blinking taillights) directly informs safe decision-making. This contribution sits at a rich confluence of 3D scene understanding, part-based object modeling, and synthetic data generation, pushing forward the broader goal of building visual AI systems with human-like granular object comprehension. With 12 citations, the work has begun attracting attention from the computer vision and robotics communities. Miao's research represents an important step toward more robust, interaction-aware perception systems capable of operating reliably in complex, real-world environments.