Multiple View Geometry in Computer Vision
Multiple View Geometry in Computer Vision
IEEE Transactions on Pattern Analysis and Machine Intelligence
| Hi-index | 0.00 |
Photos often exhibit many perspectively slanted planar parts like buildings, roads, or shop signages. We present an intuitive user interface to quickly select areas in photos using perspective dragging. When loading an image, our system preliminary hierarchically segments regions that are potentially interpreted as perspective quads. During an interactive image editing session, whenever a user selects an area by first clicking a corner and then dragging the diagonally opposite corner, the system proposes a perspective sub-rectangle matching the dragged diagonal defined according to the segmented quad containing this diagonal. We illustrate the perspective click-and-drag selection with common image editing operations.