Fan, Jialue

Email: jialue DOT fan AT u DOT northwestern DOT edu

Executable Code

Scribble Tracker executable code achieves target boundary in object tracking [PAMI'12 paper | ECCV'10 paper]

Discriminative Attention Tracker executable code handles spatial distraction, and it achieves optimal sub-regions in object tracking [ECCV'10 paper]


Note: IEEE TPAMI has a very high journal impact factor that is ranked No.1 in all IEEE publication, No.1 in both electrical engineering and artificial intelligence, and among top 3 in all computer science journals [Quote].

J. Fan, X. Shen, and Y. Wu. What Are We Tracking: A Unified Approach of Tracking and Recognition. IEEE Transactions on Image Processing (TIP), vol.22, no.2, pp.549-560, February 2013 [PDF | bibtex] We argue that high-level semantic correspondences are indispensable to make tracking more reliable. So we actively recognize the target during tracking, and the recognition result is fed back to help improve tracking.

J. Fan, X. Shen, and Y. Wu. Scribble Tracker: A Matting-based Approach for Robust Tracking. IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), vol.34, no.8, pp.1633-1644, August 2012 [PDF | Project Page | PPT | Executable Code | bibtex] We achieve the target boundary, instead of the rectangular bounding box. The proposed method alleviates drift, and it naturally handles scale change, out of plane rotation, fast motion, large deformation, and severe occlusion. The automatic generated scribble trimap is very informative in tracking, and it provides the rough location of the target. So this scribble trimap can be a complementary tool for other video tracking methods. The conference version appeared in ECCV 2010: J. Fan, X. Shen, and Y. Wu, Closed-Loop Adaptation for Robust Tracking. [PDF | bibtex] (acceptance rate: overall 27.6%, “motion and tracking” area 22.7%)

J. Fan, Y. Wu, and S. Dai. Discriminative Spatial Attention for Robust Tracking. European Conference on Computer Vision (ECCV) 2010. [PDF | Extended TR | Project Page | PPT | Executable Code | bibtex] We study spatial distraction, which is one of the major reasons leading to tracking failure. We define the concept of discriminative margin to measure the region discrimination, and design an efficient method to select discriminative reigions for object tracking. (acceptance rate: overall 27.6%, “motion and tracking” area 22.7%)

J. Fan, W. Xu, Y. Wu, and Y. Gong. Human Tracking Using Convolutional Neural Networks. IEEE Transactions on Neural Networks (TNN), vol.21, no.10, pp.1610-1623, Oct. 2010. [PDF | bibtex] It is the first work in the world of using deep learning for human tracking.

Y. Wu and J. Fan. Contextual Flow. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2009 (Oral). [PDF | Project Page | PPT | bibtex] We leap from brightness constancy to context constancy, and present a new approach that incorporates contexts to constrain motion estimation for target tracking. (acceptance rate for oral presentation: overall 4.1%, “motion and tracking” area 5.7%)

Disclaimer: Electronic papers shared on this website are for fast dissemination of research work. The copyrights of the papers belong to the corresponding publishers.