SpaceJAM: a Lightweight and Regularization-free Method for Fast Joint Alignment of Images

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Barel, Nir, Weber, Ron Shapira, Mualem, Nir, Finder, Shahaf E., Freifeld, Oren
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913848291753984
author Barel, Nir
Weber, Ron Shapira
Mualem, Nir
Finder, Shahaf E.
Freifeld, Oren
author_facet Barel, Nir
Weber, Ron Shapira
Mualem, Nir
Finder, Shahaf E.
Freifeld, Oren
contents The unsupervised task of Joint Alignment (JA) of images is beset by challenges such as high complexity, geometric distortions, and convergence to poor local or even global optima. Although Vision Transformers (ViT) have recently provided valuable features for JA, they fall short of fully addressing these issues. Consequently, researchers frequently depend on expensive models and numerous regularization terms, resulting in long training times and challenging hyperparameter tuning. We introduce the Spatial Joint Alignment Model (SpaceJAM), a novel approach that addresses the JA task with efficiency and simplicity. SpaceJAM leverages a compact architecture with only 16K trainable parameters and uniquely operates without the need for regularization or atlas maintenance. Evaluations on SPair-71K and CUB datasets demonstrate that SpaceJAM matches the alignment capabilities of existing methods while significantly reducing computational demands and achieving at least a 10x speedup. SpaceJAM sets a new standard for rapid and effective image alignment, making the process more accessible and efficient. Our code is available at: https://bgu-cs-vil.github.io/SpaceJAM/.
format Preprint
id arxiv_https___arxiv_org_abs_2407_11850
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SpaceJAM: a Lightweight and Regularization-free Method for Fast Joint Alignment of Images
Barel, Nir
Weber, Ron Shapira
Mualem, Nir
Finder, Shahaf E.
Freifeld, Oren
Computer Vision and Pattern Recognition
The unsupervised task of Joint Alignment (JA) of images is beset by challenges such as high complexity, geometric distortions, and convergence to poor local or even global optima. Although Vision Transformers (ViT) have recently provided valuable features for JA, they fall short of fully addressing these issues. Consequently, researchers frequently depend on expensive models and numerous regularization terms, resulting in long training times and challenging hyperparameter tuning. We introduce the Spatial Joint Alignment Model (SpaceJAM), a novel approach that addresses the JA task with efficiency and simplicity. SpaceJAM leverages a compact architecture with only 16K trainable parameters and uniquely operates without the need for regularization or atlas maintenance. Evaluations on SPair-71K and CUB datasets demonstrate that SpaceJAM matches the alignment capabilities of existing methods while significantly reducing computational demands and achieving at least a 10x speedup. SpaceJAM sets a new standard for rapid and effective image alignment, making the process more accessible and efficient. Our code is available at: https://bgu-cs-vil.github.io/SpaceJAM/.
title SpaceJAM: a Lightweight and Regularization-free Method for Fast Joint Alignment of Images
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.11850