Sign Language Recognition has become the active area of research nowadays. This paper describes a novel approach towards a system to recognize the different alphabets of Indian Sign Language in video sequence automatically. The proposed system comprises of four major modules: Data Acquisition, Pre-processing, Feature Extraction and Classification. Pre-processing stage involves Skin Filtering and histogram matching after which Eigen vector based Feature Extraction and Eigen value weighted Euclidean distance based Classification Technique was used. 24 different alphabets were considered in this paper where 96% recognition rate was obtained. Keywords: Eigen value, Eigen vector, Euclidean Distance (ED),Human Computer Interaction, Indian Sign Language (ISL), Skin Filtering. Cite as:Joyeeta Singh, Karen Das "Automatic Indian Sign Language Recognition for Continuous Video Sequence", ADBU J.Engg.Tech., 2(1)(2015) 0021105(5pp)