Research Experience

 
 
 
 
 

Leader of Lightning Location Project

AI Group of Dian Group, Huazhong University of Science and Technology

Dec. 2019 – Present Wuhan, China
Research Content: Parallel Computing, NVIDIA CUDA.

  • Advisors: Yayu Gao, Chengwei Zhang and Xiaojun Hei
  • Optimize lightning location algorithm for higher accuracy, and use CUDA to parallels the algorithm.
  • Achieve over 500 times of acceleration with computing power of GPUs.
  • Work in progress.
 
 
 
 
 

Research Intern

Titan OS

Jul. 2019 – Aug. 2019 Wuhan, China
Research Content: Scene text recognition, NVIDIA TensorRT

  • Advisor: Hao Tu
  • Reference a a single-shot oiented scene text detector, TextBoxes++, to handle the task of scene text detection.
  • Train the text detector and the text recognizer jointly and make it become an end-to-end training process, hence not only boost the training speed but also bring promotion in accuracy of detection as well as recognition.
  • Employ NVIDIA TensorRT, which is a platform for high-performance deep learning inference, to increase the inference speed at the cost of lower the accuracy a little bit.
 
 
 
 
 

Core member of Intelligent Transportation Project

AI Group of Dian Group, Huazhong University of Science and Technology

Mar. 2019 – Present Wuhan, China
Research Content: Object Detection, Semantic Segmentation, Sequence Image Classification.

  • Advisors: Yayu Gao, Chengwei Zhang and Xiaojun Hei
  • Integrate latest methods in Computer Vision to solve the problem of judging vehicles violate traffic rules.
  • Employ object detector e.g. YOLO and Mask RCNN to detect the vehicles and their license plates in given images.
  • Propose a branched, spatio-temporal convolution neural network, named SLDNet, to recognize these violation acts from photographs captured by surveillance cameras. A paper submitted to ICTC 2020.
  • Work in progress.

Contact