Gang Qian, Ph.D. - US grants

In this two-year effort, we propose to address some of the basic research issues that arise in the interpretation of video streams, simultaneously collected by a set of indoor or outdoor cameras.
Specifically, we are interested in inferring movements and activities of one or more humans using
distributed cameras. We propose to develop novel methods for detecting and tracking humans using 3D models
for body parts, and quasi-invariant recognition of activities humans are engaged in. We will make use of the recent advances made in the computational aspects of estimating posterior probability density functions using
Monte Carlo Markov Chain techniques to infer human descriptions and their activities.

Over the past decade, human motion analysis has become an important research area with critical applications. It is attracting significant research efforts in a number of disciplines, such as computer vision (vision-based motion capture, human computer interface, human identification), robotics (navigation), dance and choreography (automatic dance documentation and dance instruction), music (digital conducting) and bioengineering (rehabilitation and motor behavior). Motion analysis is a complex problem due to the 3D nature of the human body; the infinite possibilities of human movements; variability of movement execution between different people; continuously adaptive learning through feedback from and interactions with the environment; and the inherent multiple levels of movement structure in terms of time, space and energy. This makes it unrealistic for a single discipline to address all aspects. Therefore, progress within each discipline moves at a slow pace.
Intellectual Merit: Arizona State University has founded the Interdisciplinary Research Environment for Motion Analysis (IREMA) initiative that integrates researchers from ten disciplines to create a holistic model for motion analysis research and education. Within IREMA, ground-breaking collaborations have been established through networks of experts, infrastructures and important applications. Using this multi-level, networked research model, the principal investigators (PIs) are able to address many critical issues of real-time motion capture, analysis and feedback. Promising results of social significance are being achieved in areas such as: Rehabilitation Research to Restore Functional Walking Ability for Spinal Cord Injured, Auditory Display Systems for Aiding Interjoint Coordination, Modeling of Human and Robotic Heuristics for Projectile Interception, Movement Based Interactive Arts Environments, Experiential learning environments for children, Extraction and Recognition of Middle and Low Level Features of Movements, Vision-based Motion Capture Using Domain Knowledge.
Using the research infrastructure (RI) grant the PIs will create a multimodal sensing and feedback environment for human motion analysis research and movement-based interactive applications. They will increase their optical motion capture system to 24 cameras, create a high-speed, high resolution 24 video camera array, complete the building of a pressure sensitive floor, acquire a new EMG system and metabolic sensing equipment, acquire required hardware to integrate optical motion capture data with EMG and 2D visual as well as metabolic sensing, increasing processing and storage capacity, creating a mobile motion capture setup, and deploying the necessary hardware and software for interactive real-time feedback. The above sensing equipment would provide high-speed, high quality, synchronous video capture of multiple views, high-precision marker-based motion capture and pressure sensing in the floor as well as on the treadmill, and audio signals. It will enable the PIs to capture human movement in its full essence. The optical motion-capture data and the pressure sensing data will be fused to provide holistic motion capture. The processed, combined data of these systems will be used to train the video based system so that robust and accurate vision-based motion-capture can be acquired using low-cost video cameras. The physiological equipment will be used in the rehabilitation projects.
Broader Impact: During this five-year project, the PIs hope to achieve major advances in motion analysis and core computer science areas: computer vision, human-computer interaction, information and data management, geometric computation, knowledge systems and robotics. These advances will have significant social impact by producing major progress in movement rehabilitation and therapy, K-12 education, security applications (gait/face recognition), and all areas involving movement training (dance, theatre, sports, firefighting, military). Finally, IREMA can serve as a new model for research and interdisciplinary collaboration, which can be adapted to other areas thereby increasing their productivity. This RI grant will establish the necessary infrastructure for paradigm shifts in motion analysis and will facilitate the overall modeling of hybrid research.

Algorithms; Bio-Informatics; Bioinformatics; Biological; CRISP; Computer Retrieval of Information on Scientific Projects Database; Construction; Data; Data Banks; Data Bases; Databank, Electronic; Databanks; Database, Electronic; Databases; Effectiveness; Facility Construction Funding Category; Funding; Goals; Grant; Institution; Investigators; Lead; Measures; Methods and Techniques; Methods, Other; NIH; National Institutes of Health; National Institutes of Health (U.S.); Operation; Operative Procedures; Operative Surgical Procedures; Pb element; Performance; Phase; Process; Range; Research; Research Personnel; Research Resources; Researchers; Resources; Scanning; Source; Staging; Structure; Surgical; Surgical Interventions; Surgical Procedure; Techniques; Trees; United States National Institutes of Health; base; clinical data repository; clinical data warehouse; data repository; design; designing; heavy metal Pb; heavy metal lead; improved; indexing; insertion-deletion; insertion-deletion mutation; insertion/deletion; insertion/deletion mutation; novel; relational database; surgery

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. A basic operation on biological sequence databases is to locate homologous regions for a given query sequence using pair-wise alignments. Unfortunately. the dynamic programming algorithm used for sequence alignments is computationally expensive, making it prohibitive for today's rapidly-growing sequence databases. Existing alignment tools, such as FAST A and BLAST. though fast in locating candidate homologous regions, sacrifice sensitivity for efficiency -they may miss some true homologous regions in database sequences. In this project, we will develop novel indexing algorithms for large biological databases that support efficient pair-wise sequence alignments with high sensitivity. Specifically, we will incorporate widely-used substitution matrices, such as PAM and BLOSUM, into the construction algorithms of the NSP-tree (an index structure designed for sequence data) so that sequences with evolutionarily-related letters are grouped together in the structure of the NSP-tree. As a result, indexed sequence groups with unrelated letters will obtain a low score when aligned to a given query sequence, and be promptly pruned. By enhancing the pruning power of the NSP-tree, we expect that the new index-based approach will provide high sensitivity while maintaining a comparable or even higher level of efficiency than that of existing pair-wise alignment tools. The project will be conducted in four steps: 1) Developing a new dynamic programming query algorithm to handle the alignments between a query sequence and sequence groups indexed in the tree;2) Based on the substitution matrices, analyzing functionally conservative leiters in biological sequences, and creating a clustering tree that hierarchically organizes the proximity of the letters based on their evolutionary closeness;3) Designing new heuristics that incorporate the clustering tree of letters into the construction algorithms of the NSP-tree;and 4) Conducting experimental studies on the performance of the new heuristics and comparing the performance of the NSP-tree with that of the existing tools.

This award is co-funded by EPSCoR program.
This MRI project is to acquire a computing resource (35 compute nodes each having 2, 8 core Intel Xeon processors, with two nodes having GPU accelerators) that will enable a number of interesting and carefully detailed research efforts at University of Central Oklahoma. Research includes projects on particle transport, stochastic modeling, ecological modeling, bio-informatics and spread of disease. A number of well-developed broadening participation activities in education and research are also proposed.
This project is leveraging the established network of 2- and 4-year colleges for collaborations and is facilitating integration of research and education through training opportunities for students and high school teachers.