Zhaowei Cai (蔡兆伟)  
       Ph.D candidate
       Department of Electrical and Computer Engineering
       University of California, San Diego

[10/2018] The Detectron code of Cascade R-CNN is now available on github
[02/2018] Our new paper on Cascade R-CNN [arXiv] was accepted by CVPR18, and the code is now available on github
[02/2017] Our new paper on low-precision neural networks is now on arXiv and the code is available on github
[08/2016] The MS-CNN source code is released [link]! 
[02/2016] Our project on pedestrian detection was covered by press [UCSD][IEEE spectrum][Science Daily]!
[02/2016] The MFC exe for CompACT pedestrian detector is available for download [link]!
[03/2015] The DGT source code is released! Check the project page for details.

09/2013~Present  Ph.D in Department of Electrical and Computer Engineering, University of California San Diego
09/2007~07/2011  B.S. in Department of Automation, Dalian Maritime University
Research Interests
 Computer Vision, Machine Learning, Pattern Recognition, Image Processing          
 Deep Learning, Object Detection, Visual Tracking, Online Learning
Professional Experience
06/2014~Present   Research Assistant in Statistical Visual Computing Lab, advised by Nuno VasconcelosUniversity of California, San Diego
06/2018~09/2018  Research Intern in Facebook AI Research, Menlo Park
06/2016~09/2016  Research Intern in Microsoft Research, Redmond
06/2015~09/2015  Research Intern in IBM T. J. Watson Research, Yorktown Heights
07/2011~07/2013  Research Assistant in CBSR&NLPR, Institute of Automation, Chinese Academy of Sciences, Beijing
  1. Zhaowei Cai, and Nuno Vasconcelos"Cascade R-CNN: Delving into High Quality Object Detection", IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, Utah, United States, 2018 (Spotlight). [PDF][arXiv][Code][Code-Detectron][BibTex]
  2. Zhaowei Cai, Xiaodong He, Jian Sun, and Nuno Vasconcelos"Deep Learning with Low Precision by Half-wave Gaussian Quantization", IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, Hawaii, United States, 2017 (Spotlight). [PDF][arXiv][Code][BibTex]
  3. Zhaowei Cai, Quanfu Fan, Rogerio Feris, and Nuno Vasconcelos, "A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection", 14th European Conference on Computer Vision (ECCV), Amsterdam, Netherlands, 2016. [arXiv][BibTex][Code][Demo][KITTI Results]
  4. Zhaowei Cai, Mohammad Saberian, and Nuno Vasconcelos, "Learning Complexity-Aware Cascades for Deep Pedestrian Detection", IEEE International Conference on Computer Vision (ICCV), Santiago, Chile, 2015 (Oral Presentation). [PDF][MFC exe][demo][Caltech Results][KITTI Results][BibTex]
  5. Zhaowei Cai, Longyin Wen, Zhen Lei, Nuno Vasconcelos, and Stan Z. Li, "Robust Deformable and Occluded Object Tracking with Dynamic Graph", IEEE Transaction on Image Processing, 23(12), 5497-5509, December, 2014. [PDF][Project Page][BibTex]
  6. Longyin Wen, Zhaowei Cai, Zhen Lei, Dong Yi, and Stan Z. Li, "Robust Online Learned Spatio-Temporal Context Model for Visual Tracking", IEEE Transaction on Image Processing, 23(2), 785-796, February, 2014. [PDF][BibTex][code]
  7.  Zhaowei Cai, Longyin Wen, Dong Cao, Zhen Lei, Dong Yi, Stan Z. Li, “Person-Specific Face Tracking with Online Recognition”, 10th IEEE International Conference on Automatic Face and Gesture Recognition (FG), Shanghai, China, April 22-26, 2013. [PDF][BibTex]
  8.  Zhaowei Cai, Longyin Wen, Jianwei Yang, Zhen Lei, Stan Z. Li. “Structured Visual Tracking with Dynamic Graph”, 11th Asian Conference on Computer Vision (ACCV), Daejeon, Korea, November 5-9, 2012. [PDF][BibTex]
  9. Jianwei Yang, Zhaowei Cai, Longyin Wen, Zhen Lei, Guodong Guo, Stan Z. Li. “A New Projection Space for Separation of Specular and Diffuse Reflection Components in Color Images”, 11th Asian Conference on Computer Vision (ACCV), Daejeon, Korea, November 5-9, 2012. [PDF][BibTex]
  10. Longyin Wen, Zhaowei Cai, Zhen Lei, Dong Yi, Stan Z. Li. “Online Spatio-Temporal Structural Context Learning for Visual Tracking”. 12th European Conference on Computer Vision (ECCV), Firenze, Italy, October 7-13, 2012. [PDF][BibTex] 
  11. Zhaowei Cai, Qi Zhang, Longyin Wen, “No-reference Image Quality Metric Based on Visual Quality Saliency”, 5th Chinese Conference on Pattern Recognition (CCPR), Beijing, China, September 24-26, 2012. (Oral Presentation) [PDF][BibTex]
  12. Longyin Wen, Zhaowei Cai, Menglong Yang, Zhen Lei, Dong Yi, Stan Z. Li. “Online Multiple Instance Joint Model for Visual Tracking”. 9th IEEE International Conference on Advanced Video and Signal-Based Surveillance (AVSS). Beijing, China, September 18-21, 2012. [PDF][BibTex]
  1. Longyin Wen, Dawei Du, Zhaowei Cai, Zhen Lei, Ming-Ching Chang, Honggang Qi, Jongwoo Lim, Ming-Hsuan Yang, Siwei Lyu, "DETRAC: A New Benchmark and Protocol for Multi-Object Tracking", 2015[arXiv][Project Page][BibTex]
 Professional Acitivities
  Journal Reviewer: T-PAMI, T-IP, CVIU, T-MM, PR, T-CSVT, T-ITS, T-Cybernetics.
  Conference Reviewer: NeurIPS19, ICML19, ICCV19, CVPR19, NIPS18, ECCV18, ICML18, CVPR18, NIPS17, ICCV17CVPR17.

 Honors and Awards
   2013 Jacobs Scholarship, University of California, San Diego
   2008~2010 Excellent Student Scholarships, Dalian Maritime University