Jiaxin Cheng is a fourth-year Ph.D. student majoring in Computer Science. He is a graduate research assistant at ISI and working in the Visual Intelligence and Multimedia Analytics Laboratory (VIMAL). He earned M.S. in information security in 2017 from Shanghai Jiaotong University in China. He has various research experiences in computer vision, including templating matching, optical character recognition, semantic segmentation, zero-shot learning, adversarial attack detection, and style transfer. His works have been published in research conferences, including CVPR, ICCV, ACCV, and ICDAR.
* QATM: Quality Aware Template Matching for Deep Learning in CVPR 2019
* Style-Aware Normalized Loss for Improving Arbitrary Style Transfer in CVPR 2021 (Oral)
* SIGN: Spatial-information Incorporated Generative Network for Generalized Zero-shot Semantic Segmentation in ICCV 2021
* Recurrent Convolutional Strategies for Face Manipulation Detection in Videos in CVPR Workshop 2019
* A Study of Script Language Effects In Deep Neural Network Based Scene Text Detection in ICDAR 2019
* Image-to-GPS Verification Through A Bottom-Up Pattern Matching Network in ACCV 2018