Efficient Multi-Crop Saliency Partitioning for Automatic Image Cropping
This addresses a domain-specific need for applications requiring multiple disjoint crops, but it appears incremental as it builds on existing methods.
The paper tackles the problem of automatically extracting multiple non-overlapping crops from images, extending a fixed aspect ratio algorithm to do so efficiently in linear time without recomputing saliency maps.
Automatic image cropping aims to extract the most visually salient regions while preserving essential composition elements. Traditional saliency-aware cropping methods optimize a single bounding box, making them ineffective for applications requiring multiple disjoint crops. In this work, we extend the Fixed Aspect Ratio Cropping algorithm to efficiently extract multiple non-overlapping crops in linear time. Our approach dynamically adjusts attention thresholds and removes selected crops from consideration without recomputing the entire saliency map. We discuss qualitative results and introduce the potential for future datasets and benchmarks.