TY - JOUR AU - Priya Arokia AU - Anupama Patil TI - Conversion of 2D to 3D Technique for Monocular Images using Papilio One JO - Advances in Science, Technology and Engineering Systems Journal PY - 2019 VL - 4 IS - 2 SP - 299 EP - 304 DO - 10.25046/aj040238 UR - https://www.astesj.com/v04/i02/p38/ L1 - https://www.astesj.com/?sdm_process_download=1&download_id=5763 AB -

A 3D image adds realism in viewing experience and can assist in simplifying the graphical displays. A Third dimension supplement to the input can improve pattern recognition, and can be used for 3D scene reconstruction and robot navigation. Recently popularity of 3D hardware is also increased which makes it a hot topic. The production of content as 3D is not matching with its need so there is scope of improvement of these 3D contents. Monocular cues give profundity data when seeing a scene with one eye. When a spectator moves, the evident relative movement of a few stationary articles against a foundation gives indicates about their relative separation. Depth estimation from monocular cues is a difficult task because single image lacks prior information like depth information, motion information etc. In Depth using scene features depth is estimated by exploring the features like shape, edges, color, texture and as well as an analysis of the environment of the scene that are of interest with respect to the target. Different objects have different hue and value and hence color is useful for depth estimation. Shape and texture provides disparity which is used to estimate depth. The main problem in converting a 2 dimensional to 3 dimensional images using single image is that it lacks information required for reconstruction in 3D data. While doing conversion by taking different cues or combination of multiple cues from scene conversion has been done e.g. structure form shape, motion, defocus etc. But such methods work for restricted scenarios not for global scenes. For instance, outdoor algorithms worked poor for indoor algorithms. Here we have implemented automatic conversion of 2 dimensional to 3 dimensional images using monocular image which can convert global images in visually comfortable 3D image.

KW - Monocular image KW - Depth KW - Scenic feature ER -