Learning Visual Hierarchies in Hyperbolic Space for Image Retrieval

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Ziwei, Ramasinghe, Sameera, Xu, Chenchen, Monteil, Julien, Bazzani, Loris, Ajanthan, Thalaiyasingam
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915709134569472
author Wang, Ziwei
Ramasinghe, Sameera
Xu, Chenchen
Monteil, Julien
Bazzani, Loris
Ajanthan, Thalaiyasingam
author_facet Wang, Ziwei
Ramasinghe, Sameera
Xu, Chenchen
Monteil, Julien
Bazzani, Loris
Ajanthan, Thalaiyasingam
contents Structuring latent representations in a hierarchical manner enables models to learn patterns at multiple levels of abstraction. However, most prevalent image understanding models focus on visual similarity, and learning visual hierarchies is relatively unexplored. In this work, for the first time, we introduce a learning paradigm that can encode user-defined multi-level complex visual hierarchies in hyperbolic space without requiring explicit hierarchical labels. As a concrete example, first, we define a part-based image hierarchy using object-level annotations within and across images. Then, we introduce an approach to enforce the hierarchy using contrastive loss with pairwise entailment metrics. Finally, we discuss new evaluation metrics to effectively measure hierarchical image retrieval. Encoding these complex relationships ensures that the learned representations capture semantic and structural information that transcends mere visual similarity. Experiments in part-based image retrieval show significant improvements in hierarchical retrieval tasks, demonstrating the capability of our model in capturing visual hierarchies.
format Preprint
id arxiv_https___arxiv_org_abs_2411_17490
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Learning Visual Hierarchies in Hyperbolic Space for Image Retrieval
Wang, Ziwei
Ramasinghe, Sameera
Xu, Chenchen
Monteil, Julien
Bazzani, Loris
Ajanthan, Thalaiyasingam
Computer Vision and Pattern Recognition
Structuring latent representations in a hierarchical manner enables models to learn patterns at multiple levels of abstraction. However, most prevalent image understanding models focus on visual similarity, and learning visual hierarchies is relatively unexplored. In this work, for the first time, we introduce a learning paradigm that can encode user-defined multi-level complex visual hierarchies in hyperbolic space without requiring explicit hierarchical labels. As a concrete example, first, we define a part-based image hierarchy using object-level annotations within and across images. Then, we introduce an approach to enforce the hierarchy using contrastive loss with pairwise entailment metrics. Finally, we discuss new evaluation metrics to effectively measure hierarchical image retrieval. Encoding these complex relationships ensures that the learned representations capture semantic and structural information that transcends mere visual similarity. Experiments in part-based image retrieval show significant improvements in hierarchical retrieval tasks, demonstrating the capability of our model in capturing visual hierarchies.
title Learning Visual Hierarchies in Hyperbolic Space for Image Retrieval
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2411.17490