Efficient Multi-Crop Saliency Partitioning for Automatic Image Cropping

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hamara, Andrew, Freeman, Andrew C.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916814829649920
author Hamara, Andrew
Freeman, Andrew C.
author_facet Hamara, Andrew
Freeman, Andrew C.
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2506_22814
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Efficient Multi-Crop Saliency Partitioning for Automatic Image Cropping
Hamara, Andrew
Freeman, Andrew C.
Computer Vision and Pattern Recognition
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.
title Efficient Multi-Crop Saliency Partitioning for Automatic Image Cropping
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.22814