Projects
-
Audio driven virtual human generation
Take audio and a reference image as input, generate lip-syned video.
-
Video Super Resolution
Improve old movies' visual quality
-
Automatic trailer generation using AI
Take a long video/movie and interested key words as inputs, automatically generate trailers
-
Foreground segmentation
Segment foreground (human) and background first, and then apply style transfer separately.
Blend foreground and background together after style transfer.
-
-
-
-
-
-
-
-
-
Publications
-
Normalized Avatar Synthesis Using StyleGAN and Perceptual Refinement
Huiwen Luo, Koki Nagano, Han-wei Kung, Qingguo Xu, Zejian Wang, Lingyu Wei, Liwen Hu, Hao Li
IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2021. (PDF)
-
Mask-off: Synthesizing Face Images in the Presence of Head-mounted Displays
Yajie Zhao, Qingguo Xu, Weikai Chen, Chao Du, Jun Xing, Xinyu Huang, Ruigang Yang
IEEE Conference on Virtual Reality and 3D User Interfaces (VR), 2019. (PDF)
-
Littlehelper: An Augmented Reality Glass Application to Assist Individuals with Autism in Job Interview
Qingguo Xu, Sen-ching Samson Cheung, Neelkamal Soares
Asia Pacific Signal and Information Processing Association (APSIPA), 2015. (PDF)
-
An Advanced Methodology for Measuring and Characterizing Software Aging
Pengfei Zheng, Qingguo Xu, Yong Qi
International Workshop on Software Aging and Rejuvenation (WoSAR), 2012. (PDF)