Research Domains

New methods have been proposed to use visual object trajectories in surveillance videos for various applications. Trajectory extraction, representation, clustering and visualization, event detection, abnormality detection, trajectory-based retrieval of surveillance videos, behavior/scene understanding, and summarization / synopsis generation are some of the key contributions made in this domain from our lab.

1. A. Ahmed, D. P. Dogra, S. Kar, R. Patnaik, S. Lee, H. Choi, I. Kim. Query-based Video Synopsis using Trajectory Clustering, IEEE Transactions on Intelligent Transportation Systems, DOI: 10.1109/TITS.2019.2929618, 2019.
2. A. Ahmed, D. P. Dogra, S. Kar, P.P Roy. Clustering, Trajectory-based Surveillance Analysis: A Survey, IEEE Transactions on Circuits and Systems for Video Technology, DOI: 10.1109/TCSVT.2018.2857489, 2018.

Traffic violations, congestion, and accidents are on the rise with the rapid increase in urbanization. For efficient traffic management, it is essential to build Intelligent Transportation Systems (ITS) to ensure traffic efficiency, safety, law enforcement, etc. Recent advancements in computer and machine vision techniques have opened new opportunities to understand and analyze road traffic using unsupervised methods.

1. K K Santhosh, D. P. Dogra, P. P. Roy, Temporal unknown incremental clustering model for analysis of traffic surveillance videos, IEEE Transactions on Intelligent Transportation Systems, 20(5):1762-1773, 2019. 2018.
2. K. K. Santhosh, D. P. Dogra, P. Roy, B. B. Chaudhuri, Trajectory-based Scene Understanding using Dirichlet Process Mixture Model, IEEE Transactions on Cybernetics, DOI: 10.1109/TCYB.2019.2931139, 2019.

In recent years, AR guided education has gained its popularity due to various reasons including availability of low-cost equipment, improvement in sensing technology, and open source development platforms. This project aims to develop Augmented Reality (AR) systems that can be used as assistive tools by the instructors to deliver contents interactively during the classroom teaching for kindergarten, high school or technical higher education students.

1. A.K. Dash, S. K. Behera, D. P. Dogra, P. P. Roy. Designing of Marker-based Augmented Reality Learning Environment for Kids Using Convolutional Neural Network Architecture, Displays, 55:46-54, 2018.

Signature recognition is identifying the signature's owner, whereas signature verification is to investigate whether the signature is genuine or forgery. While both verification and recognition are important in the field of forensic science, the former is of special importance for banks and credit card companies. Feature extraction plays an important role in the signature classification process. It is the main building block of signature verification and recognition processes. This research work analyzes various aspects of designing new features that can produce better results even consuming substantially lesser time as compared to the existing features.

1. S K Behera, S Bhoi, D P Dogra, P P Roy. Robustness Analysis of Leap Motion Sensor Guided Air Authentication System, IEEE Transactions on Consumer Electronics, 64(2):171-179, 2018.

Developing anomaly event detection in surveillance videos is an uphill task, as abnormal behavior is subjective. So there is a need for developing a generalized model to detect such abnormal events. In the current scenario, the development of Deep Learning models can detect and localize anomaly events with greater accuracy than the existing state-of-the-art techniques. This research work focuses on developing such Deep Learning models for robust and efficient anomaly detection in videos.

Estimation of biological vitals of human being such as pulse rate using computer-vision and machine-learning techniques.

1. A. Sikdar, S. K. Behera, D. P. Dogra and H. Bhaskar, "Contactless vision-based pulse rate detection of Infants Under Neurological Examinations," 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC), Milan, Italy, 2015, pp. 650-653.
2. A. Sikdar, S. K. Behera and D. P. Dogra, "Computer-Vision-Guided Human Pulse Rate Estimation: A Review," IEEE Reviews in Biomedical Engineering, vol. 9, pp. 91-105, 2016.