|
Juan Carlos Niebles |
I am currently visiting Stanford University.
Office: Room 240 Gates Computer Science Building.
Email: jniebles [at] princeton [dot] edu
My graduate studies have been partly sponsored by The Fulbright Program, Colciencias and Universidad del Norte (at Barranquilla, Colombia).
Research
My research work is in computer vision, advised by Prof. Fei-Fei Li.
The main topic of my research is video analysis for automatic detection and recognition of human motion. This includes computer algorithms that can automatically find people in real-life video sequences and reason about what they are doing. My work usually involves techniques from image processing, statistics and machine learning.
I have also worked in other topics which include object recognition, embedded systems and robotics.
I was part of the UIUC-Princeton Team that won the First Place (software league) in the 1st Sematic Robot Vision Challenge at AAAI-07. You can visit our Team Page; we were featured in The Wire, the ECE Illinois Weekly Newsletter; and also in the Princeton CS Department news.
I was an organizer for BAVM2009: Bay Area Vision Meeting on Image and Video Understanding; Stanford, CA. August 14th, 2009. Read More
Selected Publications
- Juan Carlos Niebles, Bohyung Han, Andras Ferencz and Li Fei-Fei. Extracting Moving People from Internet Videos. To appear in the 10th European Conference on Computer Vision, Marseilles, France, 2008. PDF. Project Page. Code.
- Juan Carlos Niebles, Hongcheng Wang and Li Fei-Fei. Unsupervised Learning of Human Action Categories Using Spatial-Temporal Words. International Journal of Computer Vision, 2008. Springer Link. PDF. Project Page.
- Silvio Savarese, Andrey Del Pozo, Juan Carlos Niebles and Li Fei-Fei. Spatial-Temporal Correlatons for Unsupervised Action Classification, IEEE Workshop on Motion and Video Computing, Copper Mountain, Colorado January 8-9, 2008. PDF.
- Juan Carlos Niebles and Li Fei-Fei. A Hierarchical Model of Shape and Appearance for Human Action Classification. IEEE Computer Vision and Pattern Recognition (CVPR), Minneapolis, 2007. PDF.



