Jing Liao

Hong Kong University of Science and Technology, City University of Hong Kong

Papers

Total Citations

H-Index

About

Jing Liao is a leading researcher at the forefront of computer graphics and 3D computer vision, with a particular focus on generative models and novel view synthesis. Her work bridges the gap between traditional image-based rendering and modern deep learning, enabling the creation of immersive visual experiences. One of her most notable contributions is the development of the first technique for creating gigapixel panorama video loops, a combinatorial optimization method that synthesizes wide-angle, high-resolution looping videos from a grid of registered footage. This foundational work has garnered 16 citations and opened new possibilities for dynamic scene capture. More recently, Liao has advanced the field of 3D generation with her work on "CAD: Photorealistic 3D Generation via Adversarial Distillation," which addresses the growing demand for high-quality 3D data in AR/VR, robotics, and gaming. By refining the Score Distillation Sampling (SDS) algorithm, she has achieved photorealistic 3D object synthesis, a breakthrough that is already influencing next-generation generative pipelines. With her innovative approaches to both panoramic video and 3D content creation, Jing Liao continues to shape how we capture, generate, and interact with visual data.