Image Collage on Arbitrary Shape via Shape-Aware Slicing and Optimization

TL;DR

This work tackles the challenge of generating informative and aesthetically pleasing image collages for arbitrary, irregular shapes. It introduces Shape-Aware Slicing (SAS) built on a Medial Axis–based Binary Slicing Tree (MABST) to partition shapes into balanced convex patches, and an Image Content Analyzing module that assigns salient image regions via bounding boxes $Sb$ and importance ranks. An optimization framework maximizes the total salient region area across the layout while pruning the search space with a pre-configuration strategy and a triangle penalty, followed by a warping-based cell filling that preserves subject content. The approach yields shape-faithful, content-preserving collages that outperform prior methods and commercial tools on irregular shapes, with robust ablations showing the value of MAD, Axial/Crosswise directions, image assignment, and optimization. The method is demonstrated across diverse shapes and image collections, highlighting practical applicability and providing a pathway for future semantics-aware shaping and more flexible visualizations.

Abstract

Image collage is a very useful tool for visualizing an image collection. Most of the existing methods and commercial applications for generating image collages are designed on simple shapes, such as rectangular and circular layouts. This greatly limits the use of image collages in some artistic and creative settings. Although there are some methods that can generate irregularly-shaped image collages, they often suffer from severe image overlapping and excessive blank space. This prevents such methods from being effective information communication tools. In this paper, we present a shape slicing algorithm and an optimization scheme that can create image collages of arbitrary shapes in an informative and visually pleasing manner given an input shape and an image collection. To overcome the challenge of irregular shapes, we propose a novel algorithm, called Shape-Aware Slicing, which partitions the input shape into cells based on medial axis and binary slicing tree. Shape-Aware Slicing, which is designed specifically for irregular shapes, takes human perception and shape structure into account to generate visually pleasing partitions. Then, the layout is optimized by analyzing input images with the goal of maximizing the total salient regions of the images. To evaluate our method, we conduct extensive experiments and compare our results against previous work. The evaluations show that our proposed algorithm can efficiently arrange image collections on irregular shapes and create visually superior results than prior work and existing commercial tools.

Figures (22)

• Figure 1: Images collages are generated by our proposed method. With the same image collection, we collage it to various irregular shapes.
• Figure 2: Example collages are generated by previous work and commercial applications. The black silhouette above the collage is the input shape.
• Figure 3: System framework
• Figure 4: Illustration of Step one through five of MAD. (a) and (b) are the interior medial axis and exterior medial axis.
• Figure 5: (a) Before discarding cuts, (b) Discarding cuts that have higher protrusion strength