Compact Hypercube Embeddings for Fast Text-based Wildlife Observation Retrieval

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Moummad, Ilyass, Miron, Marius, Robinson, David, Zaher, Kawtar, Goëau, Hervé, Pietquin, Olivier, Bonnet, Pierre, Chemla, Emmanuel, Geist, Matthieu, Joly, Alexis
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866918429407051776
author Moummad, Ilyass
Miron, Marius
Robinson, David
Zaher, Kawtar
Goëau, Hervé
Pietquin, Olivier
Bonnet, Pierre
Chemla, Emmanuel
Geist, Matthieu
Joly, Alexis
author_facet Moummad, Ilyass
Miron, Marius
Robinson, David
Zaher, Kawtar
Goëau, Hervé
Pietquin, Olivier
Bonnet, Pierre
Chemla, Emmanuel
Geist, Matthieu
Joly, Alexis
contents Large-scale biodiversity monitoring platforms increasingly rely on multimodal wildlife observations. While recent foundation models enable rich semantic representations across vision, audio, and language, retrieving relevant observations from massive archives remains challenging due to the computational cost of high-dimensional similarity search. In this work, we introduce compact hypercube embeddings for fast text-based wildlife observation retrieval, a framework that enables efficient text-based search over large-scale wildlife image and audio databases using compact binary representations. Building on the cross-view code alignment hashing framework, we extend lightweight hashing beyond a single-modality setup to align natural language descriptions with visual or acoustic observations in a shared Hamming space. Our approach leverages pretrained wildlife foundation models, including BioCLIP and BioLingual, and adapts them efficiently for hashing using parameter-efficient fine-tuning. We evaluate our method on large-scale benchmarks, including iNaturalist2024 for text-to-image retrieval and iNatSounds2024 for text-to-audio retrieval, as well as multiple soundscape datasets to assess robustness under domain shift. Results show that retrieval using discrete hypercube embeddings achieves competitive, and in several cases superior, performance compared to continuous embeddings, while drastically reducing memory and search cost. Moreover, we observe that the hashing objective consistently improves the underlying encoder representations, leading to stronger retrieval and zero-shot generalization. These results demonstrate that binary, language-based retrieval enables scalable and efficient search over large wildlife archives for biodiversity monitoring systems.
format Preprint
id arxiv_https___arxiv_org_abs_2601_22783
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Compact Hypercube Embeddings for Fast Text-based Wildlife Observation Retrieval
Moummad, Ilyass
Miron, Marius
Robinson, David
Zaher, Kawtar
Goëau, Hervé
Pietquin, Olivier
Bonnet, Pierre
Chemla, Emmanuel
Geist, Matthieu
Joly, Alexis
Information Retrieval
Computer Vision and Pattern Recognition
Machine Learning
Multimedia
Sound
Large-scale biodiversity monitoring platforms increasingly rely on multimodal wildlife observations. While recent foundation models enable rich semantic representations across vision, audio, and language, retrieving relevant observations from massive archives remains challenging due to the computational cost of high-dimensional similarity search. In this work, we introduce compact hypercube embeddings for fast text-based wildlife observation retrieval, a framework that enables efficient text-based search over large-scale wildlife image and audio databases using compact binary representations. Building on the cross-view code alignment hashing framework, we extend lightweight hashing beyond a single-modality setup to align natural language descriptions with visual or acoustic observations in a shared Hamming space. Our approach leverages pretrained wildlife foundation models, including BioCLIP and BioLingual, and adapts them efficiently for hashing using parameter-efficient fine-tuning. We evaluate our method on large-scale benchmarks, including iNaturalist2024 for text-to-image retrieval and iNatSounds2024 for text-to-audio retrieval, as well as multiple soundscape datasets to assess robustness under domain shift. Results show that retrieval using discrete hypercube embeddings achieves competitive, and in several cases superior, performance compared to continuous embeddings, while drastically reducing memory and search cost. Moreover, we observe that the hashing objective consistently improves the underlying encoder representations, leading to stronger retrieval and zero-shot generalization. These results demonstrate that binary, language-based retrieval enables scalable and efficient search over large wildlife archives for biodiversity monitoring systems.
title Compact Hypercube Embeddings for Fast Text-based Wildlife Observation Retrieval
topic Information Retrieval
Computer Vision and Pattern Recognition
Machine Learning
Multimedia
Sound
url https://arxiv.org/abs/2601.22783