Ok, let’s see if I have this straight: You draw/sketch some objects on a canvas. You label those objects. Then the application goes out and does a search on those objects and returns images that best fits them.
Then the application composes a new image based on those found images.
Check. And crazy.
We present a system that composes a realistic picture from a simple freehand sketch annotated with text labels. The composed picture is generated by seamlessly stitching several photographs in agreement with the sketch and text labels; these are found by searching the Internet. Although online image search generates many inappropriate results, our system is able to automatically select suitable photographs to generate a high quality composition, using a filtering scheme to exclude undesirable images. We also provide a novel image blending algorithm to allow seamless image composition. Each blending result is given a numeric score, allowing us to find an optimal combination of discovered images. Experimental results show the method is very successful; we also evaluate our system using the results from two user studies.
Some are calling this fake but I don’t think so. The presentation is probably prettified but projects like Photosynth from Microsoft and the University of Washington show that this is doable… or at least imaginable.
