Robust Human Action Recognition Using AREI Features and Trajectory Analysis from Silhouette Image Sequence

ABSTRACT

In the present work, we have proposed an efficient approach for human action recognition (HAR) from silhouette image sequence in videos. The efficiency of the approach lies in feature extraction and action classification. The proposed approach includes scale-shift normalization and distorted silhouette removal for the extraction of newly introduced spatiotemporal features coined as active region energy feature (AREF), and trajectory analysis. On the other hand, classification is done using hierarchical structure. An active region is the changing region in two consecutive silhouettes to accomplish the action. The AREF is estimated using active region energy image (AREI), which embraces the energies of active regions. The higher values in the AREI signify the more activeness (changing) of that region across the silhouette sequences; i.e. the region is used more (active) to complete the action. The silhouette normalization technique makes the feature extraction more robust and scale invariant. Also, the proposed approach uses a low-dimensional feature vector, which makes the whole procedure effective regarding cost in terms of timing requirement. The experimental results on publicly available Weizmann and MuHVAi data-sets clearly validate the efficiency of the proposed technique on that of the related research work regarding accuracy in the human action detection.

KEYWORDS:

• Active region energy image (AREI)
• Distorted object removal
• Human action recognition (HAR)
• Normalize foreground
• Scale-shift normalization
• Silhouette images
• Trajectory analysis (TA)

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research work is funded by Department of Science and Technology, Government of India, through INSPIRE project, and is supported by the TEQIP phase-II project of the University of Calcutta [Fellow Id: IF10163; Ref: DST/INSPIRE Fellowship/20].

Notes on contributors

Satyabrata Maity

Satyabrata Maity is presently working as a research scholar in Computer Vision Lab, A.K.C.S.I. T, CU. Prior to this, he received his MTech degree in IT from A.K.C.S.I.T, CU, and MSc degree in computer science from Vidyasagar University in 2009 and 2006, respectively. Besides this, he was a faculty in Meghnad Saha Institute of Technology during 2010. His research interests include image and video processing, computer vision, and machine learning.

E-mail: [email protected]

Amlan Chakrabarti

Amlan Chakrabarti is at present the dean faculty of Engineering and Technology and the director of the A.K. Choudhury School of Information Technology, University of Calcutta. He received his MTech degree in radiophysics and electronics (2001) from the University of Calcutta, and has done his doctoral research on quantum computing and related VLSI design at Indian Statistical Institute, Kolkata, 2004–2008. He was a post-doctoral fellow at the prestigious School of Engineering, Princeton University, NJ, USA, during 2011–2012. He was also a design engineer at Nikkel India (OrCAD Inc. subsidiary), Bangalore, from 1999 to 2000. He is the recipient of BOYSCAST fellowship award in the area of Engineering Science from the Department of Science and Technology, Government of India in 2011, INSA Visiting Scientist Fellowship in 2014, and JSPS Invitation Fellowship in 2016. He has published more than 100 research papers in referred journals and conferences. He is a Senior Member of IEEE, Joint Secretary of IEEE CEDA –India Chapter, and life member of Computer Society of India. His research interests are quantum computing, VLSI and embedded system design, video and image processing algorithms, and data analytics.

E-mail: [email protected]

Debotosh Bhattacharjee

Debotosh Bhattacharjee received the Master of Computer Science and Engineering and the PhD (Engineering) degrees from Jadavpur University, India, in 1997 and 2004, respectively. He was associated with different institutes in various capacities until March 2007. After that, he joined his Alma Mater, Jadavpur University, as a reader in the Department of Computer Science and Engineering. Currently, he is working as a full professor in the same department. He has successfully discharged the duties of coordinating an evening master's course, MTech degree in Computer Technology from August 2008 to August 2016. His research interests pertain to the applications of machine learning techniques for face recognition, gait analysis, hand geometry recognition, histopathological image analysis, and biomedical imaging. He has authored or co-authored more than 220 journals, conference publications, including several book chapters in the areas of Biometrics and Medical Image Processing. Two US patents have been granted on his works. For post-doctoral research, Dr Bhattacharjee has visited different universities abroad like the University of Twente, The Netherlands; Instituto Superior Técnico, Lisbon, Portugal; University of Bologna, Italy; ITMO National Research University, St. Petersburg, Russia; University of Ljubljana, Slovenia; and Heidelberg University, Germany. He is a life member of Indian Society for Technical Education (ISTE, New Delhi), Indian Unit for Pattern Recognition and Artificial Intelligence (IUPRAI), and a senior member of IEEE (USA).

E-mail: [email protected]