CVTEAM is a laboratory of Cognitive Vision Towards Evolvable Algorithms and Models. It is attached to the State Key Laboratory of Virtual Reality Technology and Systems, School of Computer Science and Engineering, Beihang University. Currently, CVTEAM is hosted by Prof. Jia Li and the students under his supervision.
    Members of CVTEAM are now mainly now working on fundamental vision problems such as attention & saliency, image/video segmentation and semantic understanding. Intelligent tools for processing big visual data are explored as well. Some members also pay attention to integrate CV and VR. With the invaluable contribution of all CVTEAM members, we believe that something useful can be produced in the next ten years. The long-term objective of CVTEAM is to do something that can change the world with cognitive vision techniques that can evolve to work in realistic scenarios.

Publication News

  • World's largest dataset of primary object segmentation in large-scale aerial videos (APD) is publicly available in November, 2018.
    [Benchmark Page]
  • "Low-resolution Face Recognition in the Wild via Selective Knowledge Distillation", co-authored by Shiming Ge, Shengwei Zhao, Chenyu Li, Jia Li, has been accepted in IEEE TIP 2018.
  • Resources of our paper "Hierarchical Deep Cosegmentation of Primary Objects in Aerial Videos" in IEEE MultiMedia 2018 is released now. [Project Page]
  • Results of our paper "Exploring Weakly Labeled Images for Video Object Segmentation with Submodular Proposal Selection" in IEEE TIP 2018 is released now. [Results]
  • "Collaborative Annotation of Semantic Objects in Images with Multi-granularity Supervisions", co-authored by Lishi Zhang, Chenghan Fu and Jia Li, has been accepted for publication (oral) in ACM MM 2018. Code and other results are released in our Project Page.
  • A large salient object detector dataset ( denoted as XPIE ) is released. It contains 10,000 images with precise binary masks and 8,598 images that are considered to have no significant salient objects. [Dataset]

Member News

  • Our Joint Team with Baidu wins the 1st place award in WebVision 2018 Challenge on WebVision Image Classification Task.
  • Prof. Jia Li is appointed to the guest researcher of National Engineering Laboratory for Video Technology at Peking University since April 2018.
  • 2018/03/28 Anlin Zheng received the honor of Excellent Graduate of Beijing City (M.S.).
  • 2018/03/11 Lishi Zhang passed his thesis defense of Master's degree.
  • 2018/03/11 Anlin Zheng passed his thesis defense of Master's degree.