I am a PhD student in the Vision and Learning Lab at UC Merced, supervised by professor Ming-Hsuan Yang. My current research focus is on Weakly-supervised/unsupervised Learning as well as Domain Adaptation for Object Detection and Semantic Segmentation. In UC San Diego and NTU, I have also worked on 3D Vision and Reconstruction with professor Manmohan Chandraker, and Video Temporal Consistency with professor Shao-Yi Chien. My past experience spans the fields of Multimedia, Reinforcement Learning, Data Mining, System Power Optimization, and Mobile Wireless Networking. For more information, check Research!
My non-academic interests include traveling, ecological observation, photography, and badminton, a part of which is shown in Activities.
MY LATEST RESEARCH
Example-based Video Color Transfer
In this work we propose an efficient color transfer algorithm to deliver the color characteristics of the reference image to the source image or video. Also, a quality metric for video temporal consistency is proposed to evaluate the performance of existing methods.
Occlusion-aware Video Temporal Consistency
For image processing techniques like color transfer, HDR tone mapping, white balance, and dehazing, naively applying them to videos frame-by-frame often leads to temporal inconsistency, which is especially severe when occlusion occurs. We solve it with a framework of keyframe selection, temporal warping, content compensation, and post-processing.