Selected Research Projects

Medical Image Processing

We investigated novel image and video processing methods and new clinical applications for medical imaging system such as angiography, computing topography (CT), magnetic resonance (MR), etc.

Real-Time Distributed Multi-Object Tracking from A Single Camera

We proposed a method which avoids the common practice of using a complex joint state space representation and performing tedious joint data association for multiple object tracking applications. Instead, we proposed a distributed Bayesian formulation using multiple interactive trackers that requires much lower complexity for real-time tracking applications.

When the objects' observations do not interact with each other, our approach performs as multiple independent trackers. However, when the objects' observations exhibit interaction, defined as close proximity or partial and complete occlusion, we extended the conventional Bayesian tracking framework by modeling such interaction in terms of potential functions. Specifically, we modeled the interactive likelihood densities by a "gravitation attraction" versus "magnetic repulsion" scheme. Furthermore, we approximated the "2nd-order" state transition density by an ad hoc "1st-order inertia Markov chain" in a unified particle filtering implementation.

The proposed models represent the cumulative effect of virtual physical forces that objects undergo while interacting with each other. It implicitly handles the "error merge" and "object labeling" problems and thus solves the difficult object occlusion and data association problems in an innovative way. Our preliminary simulations have demonstrated that the proposed approach is far superior to other methods in both robustness and speed.

A Motion Trajectory Based Video Retrieval System

We proposed a novel motion trajectory based video retrieval system using LAMSTAR-based adaptive self organizing maps (PASOMs). The trajectories are extracted from video by a robust tracker. To reduce the high dimension of motion trajectories, we first decomposed each trajectory into sub-trajectories by using a maximum acceleration based approach. Each sub-trajectory is then modeled and coded by two different methods, polynomial curving fitting and independent component analysis. To fuse the different features of sub-trajectories for more efficient and flexible retrieval, we used PASOMs as the searching tool. Experimental results show the superior performance of the proposed approach for video retrieval comparing with prior approaches.

Multiple Target Tracking in Crowded Scenes Using Multiple Collaborative Cameras

Multiple target tracking has received tremendous attention due to its wide practical applicability in video processing and analysis applications. Most existing techniques, however, suffer from the well-known "multi-target occlusion" problem and/or immense computational cost due to its use of high-dimensional joint state representations. We proposed a distributed Bayesian framework using multiple collaborative cameras for robust and efficient multiple target tracking in crowded environments with significant and persistent occlusion. When the targets are in close proximity or present multi-target occlusions in a particular camera view, camera collaboration between different views is activated in order to handle the multi-target occlusion problem in an innovative way. Specifically, we proposed to model the camera collaboration likelihood density by using epipolar geometry with sequential Monte Carlo implementation. Experimental results have been demonstrated for both synthetic and real-world video data.

Decentralized Articulated Motion Analysis and Object Tracking from Videos

We proposed two new articulated motion analysis and object tracking approaches: Decentralized Articulated Object Tracking method and Hierarchical Articulated Object Tracking method. The first approach avoids the common practice of using a high dimensional joint state representation for articulated object tracking. Instead, we introduced a decentralized scheme and model the inter-part interaction within an innovative Bayesian framework. Specifically, we estimated the interaction density by an efficient decomposed inter-part interaction model. To handle severe self-occlusions, we further extended the first approach by modeling high-level inter-unit interaction and develop the second algorithm within a consistent hierarchical framework. Preliminary experimental results have demonstrated the superior performance of the proposed approaches on real-world videos in both robustness and speed compared with other articulated object tracking methods.

Efficient Object Tracking Using Control-Based Observer Design

Kernel-based tracking approaches have proven to be more efficient in computation compared to other tracking approaches such as particle filtering. However, existing kernel-based tracking approaches suffer from the well-known "singularity" problem. We proposed a novel object tracking framework to handle this problem by using a control-based observer design. Specifically, we formulated object tracking as a recursive inverse problem, thus unifying several approaches to video tracking, including kernel-based tracking, into a consistent theoretical framework. Moreover, we interpreted the inverse equation as a measurement process and supplement it by introducing state dynamics as a constraint. The augmented recursive inverse equation forms a state-space model, which is solved by using a control-based optimal observer. By exploiting observability theory from control engineering, we extended the current approach to kernel-based tracking and provide explicit criteria for kernel design and dynamics evaluation. The tracking performance of our approach has been demonstrated on both synthetic and real-world video data.

Automatic Multi-Head Detection and Tracking System

We presented a novel automatic system integrating head detection with particle filter for real-time multi-head tracking (MHT) in video. Distinct with the conventional particle filter which gets particles from the prior density, we proposed a novel importance function based on the up to date detection and motion observation which makes the particles more effective and helps us to achieve stable tracking by using much less particles. We also proposed a general likelihood model in the context of MHT. Different information can be fused in a principle manner to make the tracker more stable. The proposed approach can handle not only the changes of scale, lighting, zooming, and pose, but also fast motion, appearance, and hard multi-head occlusion.

More coming soon...