Welcome to the webpage of Yuan
I've been doing computer vision research in IRIS with Prof. Ram Nevatia, Viterbi School of Engineering, University of Southern California, and before that, with Prof. Haizhou Ai's lab in Dept. of Computer Science, Tsinghua Univerity, Beijing, China. Both are great vision research groups! :D
 
Go to: research | resume (pdf) |

more about me
 
Email me: yli8 AT usc DOT edu

Finding Objects in Spatio-Temporal Context

(2008)

Work after I came to USC.

I am interested in human-centered multi-category object localization and tracking.

Key Object Driven Multi-Category Object Recognition, Localization and Tracking Using Spatio-Temporal Context (what a long title... #_#)

- Yuan LI, Ram NEVATIA, ECCV 2008. [pdf]

Human Detection: Search in 3d (2008)

Human Detection by Searching in 3D Space Using Camera and Scene Knowledge.  

- Yuan LI, Bo WU, Ram NEVATIA, ICPR 2008 (oral paper). [pdf] [data_CAVIAR_INRIA(subset)] [data_PETS07(subset)]

Human Detection & Tracking (2007~2008)

Co-authored work after I came to USC.

Tracking by global data association using network flow.  

- Li ZHANG, Yuan LI, Ram NEVATIA, Global Data Association for Multi-Object Tracking
Using Network Flows. CVPR 2008 (oral paper).

Segmentation of Multiple, Partially Occluded Objects.  

- Bo WU, Ram NEVATIA, Yuan LI, Segmentation of Multiple, Partially Occluded Objects by Grouping, Merging, Assigning Part Detection Responses. CVPR 2008.

Tracking (2006~2007)

Tracking in Low Frame Rate Video / Tracking Abrupt Motion.  

Online and offline observers integrated by a cascade partial filter.

-Yuan LI, Haizhou AI, Takayoshi YAMASHITA, Shihong LAO, Masato KAWADE, Tracking in Low Frame Rate Video: A Cascade Particle Filter with Discriminative Observers of Different Lifespans, CVPR 2007 (oral paper, best student paper).

An extended journal version:

-Yuan LI, Haizhou AI, Takayoshi YAMASHITA, Shihong LAO, Masato KAWADE, Tracking in Low Frame Rate Video: A Cascade Particle Filter with Discriminative Observers of Different Lifespans, IEEE Transactions on Pattern Analysis and Machine Intelligence, to appear, 2008.

Tracking - continued (2006~2007)

Face and head tracking.  

- Yuan LI, Haizhou AI, Chang HUANG, Shihong LAO, Robust Head Tracking with Particles Based on Multiple Cues Fusion, T.S. Huang et al. (Eds.): HCI/ECCV 2006, LNCS 3979, pp.29-39, Springer-Verlag Berlin Heidelberg 2006.

Follow-up work:

- Yuan LI, Haizhou AI, Chang HUANG, Shihong LAO, Robust Head Tracking Based on a Multi-State Particle Filter, 7th IEEE International Conference, Automatic Face and Gesture Recognition, FG2006 (poster highlight), pp.335-340, Southampton, UK, April 10-12 2006.

CLEAR 2007 Evaluation (2D face detection and tracking task).  

Presentation and official results available at www.clear-evaluation.org.

- Yuan LI, Chang HUANG, Haizhou AI, Tsinghua Face Detection and Tracking for CLEAR 2007 Evaluation, CLEAR 2007 Workshop, LNCS 4625, pp. 138–147, Springer-Verlag Berlin Heidelberg 2008.

Multi-View Face Detection (2005~2007)

Faster and more robust detection of multi-view faces under various circumstances.  

A tree-structured face detector trained by Vector Boosting covers a range of +/-45 degree rotation in plane and +/-90 degree yaw extent of the face. Speed is about 10 fps on 320x240 sized images.

- Chang HUANG, Haizhou AI, Yuan LI, Shihong LAO, High Performance Rotation Invariant Multi-View Face Detection, IEEE Transactions on Pattern Analysis and Machine Intelligence,Vol.29, No.4, pp. 671-686, April 2007.

- Chang HUANG, Haizhou AI, Yuan LI, Shihong LAO, Learning Sparse Features in Granular Space for Multi-View Face Detection, 7th IEEE International Conference, Automatic Face and Gesture Recognition, FG2006 (oral paper), pp.401-406, Southampton, UK, April 10-12 2006.

- Chang HUANG, Haizhou AI, Yuan LI, Shihong LAO, Vector Boosting for Rotation Invariant Multi-View Face Detection, The IEEE International Conference on Computer Vision, ICCV-05 (oral paper), pp.446-453, Beijing, China, Oct 17-20, 2005.

Miscellaneous (2006~2007)

Detecting independent motion in crowds.  

A small piece of work inspired by Gabriel J. Brostow and Roberto Cipolla's paper “Unsupervised bayesian detection of independent motion in crowds” and Vincent Rabaud and Serge Belongie's “Counting crowded moving objects,” (both in CVPR 2006).

- Yuan Li, Haizhou Ai, Fast Detection of Independent Motion in Crowds Guided by Supervised Learning, The IEEE International Conference on Image Processing (ICIP-07), San Antonio, Texas, USA, September 16-19, 2007.

Face based video retrieval.  

I enjoyed using it to find all the shots of Rachel in Friends! :)

- Pengxu LI, Haizhou AI, Yuan LI, Chang HUANG, Video Parsing based on Head Tracking and Face Recognition. ACM International Conference on Image and Video Retrieval (CIVR-07), Amsterdam, The Netherlands, July 9-11 2007.