Action and Activity Recognition and Understanding

Action and Activity Recognition and Understanding

Details

We are in the midst of a data  revolution,  where  visual  content  has  a  protagonist  role. For instance, YouTube reports that over 100 hours of video are uploaded every minute to their servers. Uploaded content ranges from a view of astronauts walking in space to the first steps of a baby at home. Our aim is to develop novel algorithms to automatically understand and recognize human activities from this huge visual space. We hope that our research brings the possibility to develop novel applications such as: video surveillance systems capable of detecting suspicious activities, automated household assistants, monitoring performance and understanding strategy in sports, and indexing content in web services. Below are some sample publications on this topic.


Selected Publications




  A Large-Scale Video Benchmark for Human Activity Understanding

 

 

Action Recognition using Discriminative Structured Trajectory Groups



​​
 

A Topic Model Approach to Represent and Classify American Football Plays




 

Publications

  1. Fabian Caba Heilbron, Victor Castillo, Bernard Ghanem, and Juan Carlos Niebles, “ActivityNet: A Large-Scale Video Benchmark for Human Activity Understanding”, Conference on Computer Vision and Pattern Recognition (CVPR 2015)
  2. Indriyati Atmosukarto, Narendra Ahuja, Bernard Ghanem, "Action Recognition using Discriminative Structured Trajectory Groups", Winter Conference on Applications of Computer Vision (WACV 2015)
  3. Jagannadan Varadarajan, Indriyati Atmosukarto, Shaunak Ahuja, Bernard Ghanem, Narendra Ahuja, "Automated Playbook Generation in Football through Videos ", MIT Sloan Sports Analytics Conference 
  4. Fabian Caba Heilbron, Ali Thabet, Juan Carlos Niebles, Bernard Ghanem, "Camera Motion and Surrounding Scene Appearance as Context for Action Recognition", Asian Conference on Computer Vision (ACCV 2014)
  5. Jagannadan Varadarajan, Indriyati Atmosukarto, Shaunak Ahuja, Bernard Ghanem, Narendra Ahuja, "A Topic Model Approach to Represent and Classify American Football Plays", British Machine Vision Conference (BMVC2013)
  6. Indriyati Atmosukarto, Bernard Ghanem, Shaunak Ahuja, Karthik Muthuswamy, Narendra Ahuja, "Automatic Recognition of Offensive Team Formation in American Football Plays", IEEE International Workshop on Computer Vision in Sports (CVsports), 2013 
  7. Indriyati Atmosukarto, Bernard Ghanem, Narendra Ahuja, "Trajectory-based Fisher Kernel Representation for Action Recognition in Videos", International Conference on Pattern Recognition (ICPR2012)