Friday, June 23, 2017 - 16:45 to 18:00
Sparse coded handcrafted and deep features for colon capsule video summarization
Abstract—Capsule endoscopy, which uses a wireless camera to take images of the digestive track, is emerging as an alternative to traditional wired colonoscopy. A single examination produces a sequence of approximately 50,000 frames. These sequences are manually reviewed, which is time consuming and typically takes about 45–90 minutes and requires the undivided concentration of the reviewer. In this paper, we propose a novel capsule video summarization framework using sparse coding and dictionary learning in feature space. Video frames are clustered into superframes based on power spectral density, and cluster representative frames are used for video summarization. Handcrafted and deep features that are extracted for representative frames are sparse coded using a learned dictionary. Sparse coded features are later used for training SVM classifier. The proposed method was compared to other methods, in the literature, based on sensitivity and specificity on the KID dataset. The achieved results show that our proposed framework provides robust capsule video summarization without losing informative segments.
Ahmed Mohammed's picture
Ahmed Mohammed
Norwegian University of Science and Technology (NO)
Sule Yildirim's picture
Sule Yildirim
Marius Pedersen's picture
Marius Pedersen
Oistein Hovde's picture
Oistein Hovde
Faouzi Cheikh's picture
Faouzi Cheikh