Yao Luo 骆 瑶

I'm a lecturer at the School of Computer and Electronic Information / Artificial Intelligence of Nanjing Normal University in Nanjing, China.

I did my PhD at Nanjing University of Science & Technology in Computer Science and Technology, where I was advised by Professor Jinhui Tang and Professor Jinshan Pan. Before that, I received my master's degree from Texas A&M University in Electrical Engineering and my bachelor's degree from Southeast University in Electronic Science and Technology.

Email  /  Scholar  /  Github  /  CV

Research

My current research focuses on computer vision, deep learning and spatial intelligence, especially inferring scene motion and structure from videos for enhanced visual experience, with constrained resources.

We aim to unleash the multifaceted knowledge yielded by the hierarchical views at multiple scales in a pyramid network for video frame interpolation. To this end, we present a dual-view pyramid network with an auxiliary multi-scale collaborative supervision.

We investigate the unary potentials of the characterizations to improve efficiency of video frame interpolation. To this end, we design a lightweight neural network to construct motion volumes via ensembles of offset approximations, and a spatiotemporal variance-aware loss for supervision.

We present a bi-directional pseudo-three-dimensional network to exploit the correlation between motion estimation and depth-related occlusion estimation, in both the past and future directions, synthesizing intermediate frames through a bi-directional pseudo-three-dimensional warping layer with a novel multi-task collaborative learning strategy.

We present a bi-branch network for efficient dynamic scene deblurring. The proposed network conduct heterogeneous transformations on motion and RGB content in an encoder–decoder structure with skip connections. We refine features captured by the motion branch and the color branch by incorporating a lightweight nonlocal fusion layer.

Academic Services

• Conference Reviewer: ACM MM, ICASSP, IJCNN, China MM
• Journal Reviewer: IEEE Transactions on Neural Networks and Learning Systems, Image and Vision Computing
• Volunteer: NExT++ workshop, EIPBN

Thanks Jon Barron for the website template.