Carousel: A High-Resolution Dataset for Multi-Target Automatic Image Cropping
Rafe Loya, Andrew Hamara, Benjamin Estell, Benjamin Kilpatrick, Andrew C. Freeman

TL;DR
This paper introduces Carousel, a high-resolution dataset with human-labeled multiple aesthetically appealing image crops, addressing the gap in multi-target automatic image cropping for social media applications.
Contribution
It presents a new dataset of 277 images with multiple human-labeled crops and evaluates existing single-crop models for multi-crop generation.
Findings
The dataset enables research on multi-target cropping.
Single-crop models show potential when combined with image partitioning.
Carousel dataset is publicly available for further study.
Abstract
Automatic image cropping is a method for maximizing the human-perceived quality of cropped regions in photographs. Although several works have proposed techniques for producing singular crops, little work has addressed the problem of producing multiple, distinct crops with aesthetic appeal. In this paper, we motivate the problem with a discussion on modern social media applications, introduce a dataset of 277 relevant images and human labels, and evaluate the efficacy of several single-crop models with an image partitioning algorithm as a pre-processing step. The dataset is available at https://github.com/RafeLoya/carousel.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsSmart Agriculture and AI · Advanced Neural Network Applications · Advanced Image and Video Retrieval Techniques
