Image Collage on Arbitrary Shape via Shape Partitioning and Optimization

Dong-Yi Wu, Thi-Ngoc-Hanh Le, Sheng-Yi Yao, Yun-Chen Lin, Tong-Yee Lee
book cover


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 partition scheme and a placement optimization algorithm 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 proposed a novel algorithm, called Shape Partitioning, which will first partition the input shape into cells based on medial axis and binary tree slicing. Shape Partitioning, which is designed specifically for irregular shapes, takes human perception and shape structure into account while offering different style options, namely uniform or uneven cell size. Then, the collected input images are analyzed and optimally assigned to each region of the layout by optimizing a loss function consisting of the importance and coverage ratio. To evaluate our method, we conduct extensive experiments and compare our results against previous works. The evaluations show that our proposed algorithm can efficiently arrange image collections on irregular shapes and create visually superior results than prior works and existing commercial tools.


Image collection visualization, image collage, irregular shape layout

About us

CGVSL - National Cheng-Kung University

Contact us

Results and Comparisons

Han, X., Zhang, C., Lin, W., Xu, M., Sheng, B., & Mei, T. (2015). Tree-based visualization and optimization for image collection. IEEE Transactions on Cybernetics, 46(6), 1286-1300.
Shape collage
Yu, J., Chen, L., Zhang, M., & Li, M. (2022). SoftCollage: A Differentiable Probabilistic Tree Generator for Image Collage. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 3729-3738).