_version_ 1866916413244964864
author Jain, Aditya
Cunha, Fagner
Bunsen, Michael James
Cañas, Juan Sebastián
Pasi, Léonard
Pinoy, Nathan
Helsing, Flemming
Russo, JoAnne
Botham, Marc
Sabourin, Michael
Fréchette, Jonathan
Anctil, Alexandre
Lopez, Yacksecari
Navarro, Eduardo
Pimentel, Filonila Perez
Zamora, Ana Cecilia
Silva, José Alejandro Ramirez
Gagnon, Jonathan
August, Tom
Bjerge, Kim
Segura, Alba Gomez
Bélisle, Marc
Basset, Yves
McFarland, Kent P.
Roy, David
Høye, Toke Thomas
Larrivée, Maxim
Rolnick, David
author_facet Jain, Aditya
Cunha, Fagner
Bunsen, Michael James
Cañas, Juan Sebastián
Pasi, Léonard
Pinoy, Nathan
Helsing, Flemming
Russo, JoAnne
Botham, Marc
Sabourin, Michael
Fréchette, Jonathan
Anctil, Alexandre
Lopez, Yacksecari
Navarro, Eduardo
Pimentel, Filonila Perez
Zamora, Ana Cecilia
Silva, José Alejandro Ramirez
Gagnon, Jonathan
August, Tom
Bjerge, Kim
Segura, Alba Gomez
Bélisle, Marc
Basset, Yves
McFarland, Kent P.
Roy, David
Høye, Toke Thomas
Larrivée, Maxim
Rolnick, David
contents Insects represent half of all global biodiversity, yet many of the world's insects are disappearing, with severe implications for ecosystems and agriculture. Despite this crisis, data on insect diversity and abundance remain woefully inadequate, due to the scarcity of human experts and the lack of scalable tools for monitoring. Ecologists have started to adopt camera traps to record and study insects, and have proposed computer vision algorithms as an answer for scalable data processing. However, insect monitoring in the wild poses unique challenges that have not yet been addressed within computer vision, including the combination of long-tailed data, extremely similar classes, and significant distribution shifts. We provide the first large-scale machine learning benchmarks for fine-grained insect recognition, designed to match real-world tasks faced by ecologists. Our contributions include a curated dataset of images from citizen science platforms and museums, and an expert-annotated dataset drawn from automated camera traps across multiple continents, designed to test out-of-distribution generalization under field conditions. We train and evaluate a variety of baseline algorithms and introduce a combination of data augmentation techniques that enhance generalization across geographies and hardware setups.
format Preprint
id arxiv_https___arxiv_org_abs_2406_12452
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Insect Identification in the Wild: The AMI Dataset
Jain, Aditya
Cunha, Fagner
Bunsen, Michael James
Cañas, Juan Sebastián
Pasi, Léonard
Pinoy, Nathan
Helsing, Flemming
Russo, JoAnne
Botham, Marc
Sabourin, Michael
Fréchette, Jonathan
Anctil, Alexandre
Lopez, Yacksecari
Navarro, Eduardo
Pimentel, Filonila Perez
Zamora, Ana Cecilia
Silva, José Alejandro Ramirez
Gagnon, Jonathan
August, Tom
Bjerge, Kim
Segura, Alba Gomez
Bélisle, Marc
Basset, Yves
McFarland, Kent P.
Roy, David
Høye, Toke Thomas
Larrivée, Maxim
Rolnick, David
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Insects represent half of all global biodiversity, yet many of the world's insects are disappearing, with severe implications for ecosystems and agriculture. Despite this crisis, data on insect diversity and abundance remain woefully inadequate, due to the scarcity of human experts and the lack of scalable tools for monitoring. Ecologists have started to adopt camera traps to record and study insects, and have proposed computer vision algorithms as an answer for scalable data processing. However, insect monitoring in the wild poses unique challenges that have not yet been addressed within computer vision, including the combination of long-tailed data, extremely similar classes, and significant distribution shifts. We provide the first large-scale machine learning benchmarks for fine-grained insect recognition, designed to match real-world tasks faced by ecologists. Our contributions include a curated dataset of images from citizen science platforms and museums, and an expert-annotated dataset drawn from automated camera traps across multiple continents, designed to test out-of-distribution generalization under field conditions. We train and evaluate a variety of baseline algorithms and introduce a combination of data augmentation techniques that enhance generalization across geographies and hardware setups.
title Insect Identification in the Wild: The AMI Dataset
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2406.12452