Akgül, ÖmerPenekli, H. I.Genç, Y.2018-04-122018-04-122016-11-12http://hdl.handle.net/11693/37607Date of Conference: 28 Nov.-1 Dec. 2016Conference name: 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS), 2016An existing deep learning architecture has been adapted to solve the detection problem in camera-based tracking for augmented reality (AR). A known target, in this case a planar object, is rendered under various viewing conditions including varying orientation, scale, illumination and sensor noise. The resulting corpus is used to train a convolutional neural network to match given patches in an incoming image. The results show comparable or better performance compared to state of art methods. Timing performance of the detector needs improvement but when considered in conjunction with the robust pose estimation process promising results are shown. © 2016 IEEE.EnglishAugmented realityNeural networksConvolutional neural networkDetection problemsLearning architecturesMarker detectionsMarker trackerState-of-art methodsTiming performanceViewing conditionsDeep learningConference Paper10.1109/SITIS.2016.17