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 
ORCID: https://orcid.org/0000-0002-5276-1405; Sabourin, Michael; Fréchette, Jonathan; Anctil, Alexandre; Lopez, Yacksecari; Navarro, Eduardo; Pimentel, Filonila Perez; Zamora, Ana Cecilia; Silva, José Alejandro Ramirez; Gagnon, Jonathan; August, Tom ORCID: https://orcid.org/0000-0003-1116-3385; Bjerge, Kim; Gomez Segura, Alba; Bélisle, Marc; Basset, Yves; McFarland, Kent P.; Roy, David ORCID: https://orcid.org/0000-0002-5147-0331; Høye, Toke Thomas; Larrivée, Maxim; Rolnick, David.
2024
Insect identification in the wild: the AMI dataset.
In:
Computer vision – ECCV 2024.
Cham, Switzerland, Springer Nature, 55-73.
(Lecture Notes in Computer Science, 15095, 37).
Abstract/Summary
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. The dataset is made publicly available https://github.com/RolnickLab/ami-dataset.
| Item Type: | Publication - Book Section |
|---|---|
| Digital Object Identifier (DOI): | 10.1007/978-3-031-72913-3_4 |
| UKCEH and CEH Sections/Science Areas: | Biodiversity (Science Area 2017-) Pollution (Science Area 2017-) |
| ISBN: | 978-3-031-72913-3 |
| ISSN: | 0302-9743 |
| Additional Information. Not used in RCUK Gateway to Research.: | Manuscript version available via Related URLs 'Other' link. |
| Additional Keywords: | artificial intelligence, computer networks, computer systems, computer vision, education, Human-Computer Interaction (HCI), image analysis, image coding, image processing, image reconstruction, image segmentation, learning, machine learning, object recognition, pattern recognition, reconstruction, signal processing, software engineering |
| NORA Subject Terms: | Ecology and Environment Computer Science Data and Information |
| Related URLs: | |
| Date made live: | 03 Dec 2024 13:32 +0 (UTC) |
| URI: | https://nora.nerc.ac.uk/id/eprint/538492 |
Actions (login required)
 |
View Item |
Document Downloads
Downloads for past 30 days
Downloads per month over past year
Tools
Tools
