Greedy Approach for Optimizing Image View Layout on Various Sizes of 2D UI Container

One important feature of every augmentative and alternative communication (AAC) application is the image view layout (IVL) in a user interface (VI) container, whose size is bound by the device's screen size. The IVL problem deals with arranging a set of images in a two-dimensional UI container with a definitive size. The study seeks to maximize the use of the container space to fit in as many images as possible in such a way that the number of display sets for the overall images is minimum. We consider each image a picture card containing an image or a photograph and a text label with a fixed height but varying width. The IVL problem is a special case of the 2D Bin Packing Problem, which is NP Hard. To provide approximate solutions, we propose two greedy-based heuristic algorithms, which are Order-IVL and Best-Fit-IVL. For the experimental dataset, we use 148 picture cards obtained from an AAC application called VICARA. For 30% of the experimental scenarios, our evaluations show that Order-IVL produces one extra display set than Best-Fit-IVL. Furthermore, Best-Fit-IVL results in the same number of image display sets as the optimal solutions generated by solving the integer linear programming of the IVL problem. In addition, the Best-Fit-IVL consistently produces a tidy and neater arrangement of picture cards than Order-Ivlfor all experiment scenarios.