Segmentation of An Indian Classical Dance Videos using Different Segmentation Methods
Bhavana.R.Maale1, Suvarna.Nandyal2
1Bhavana R.Maale, Computer Science & Engg, VTU, PG Centre, Kalaburagi, India.
2Dr.Suvarna Nandyal*, Computer Science & Engg, VTU,PDA College of Engg, Kalaburagi, India.
Manuscript received on November 18, 2019. | Revised Manuscript received on 27 November, 2019. | Manuscript published on December 10, 2019. | PP: 982-986 | Volume-9 Issue-2, December 2019. | Retrieval Number: I8468078919/2019©BEIESP | DOI: 10.35940/ijitee.I8468.129219
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: RVideo to frame conversion features are retrieved to categorize the actions in an Indian classical dance video dataset. The goal is to design an automatic machine learning model that identifies the moves of a dancer in a video. A video is a collection of images of specific movements, hence, features representing shapes and color can be used to interpret the dance steps. Image segmentation based features are capable of representing the shape in varying background conditions. Segmentation has become an important objective in image analysis and computer vision. To segment the images, edge detection, thresholding and region of interest are taken for this study. The proposed system performance is analyzed for total number of 50 different movements taken from Indian classical dances. Bharatanatyam, Kathak, Kuchipudi, Manipuri, Mo hiniytam Odissi, Kathakali and Satrriya in different background conditions.
Keywords: Segmentation, Edge Detection, Thresholding, Region of Interest, Feature Extraction ,K-Means clustering.
Scope of the Article: Clustering