Overview of LifeCLEF 2021: An Evaluation of Machine-Learning Based Species Identification and Species Distribution Prediction

Joly, Alexis; Goëau, Hervé; Kahl, Stefan; Picek, Lukáš; Lorieul, Titouan; Cole, Elijah; Deneu, Benjamin; Servajean, Maximillien; Durso, Andrew M.; Bolon, Isabelle; Glotin, Hervé; Planquè, Robert; Castañeda, Rafael Luis Ruiz De; Vellinga, Willem-Pier; Klinck, Holger; Denton, Tom; Eggel, Ivan; Bonnet, Pierre; Müller, Henning

doi:10.1007/978-3-030-85251-1_24

Cited by 21 publications

(14 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While their use in ecology to date has been limited, it is likely their application will grow as large multidimensional datasets become available through automated sensing technologies. For example, Transformer models for species classification and distribution prediction from image and sound recordings in the field have already begun to emerge (Conde & Turgutlu, 2021; Elliott et al, 2021; Joly et al, 2021; Reedha et al, 2022).…”

Section: From Automated Data Collection To Ecological Knowledgementioning

confidence: 99%

Towards the fully automated monitoring of ecological communities

et al. 2022

View full text Add to dashboard Cite

Ecosystems are increasingly exposed to stressors, leading to unprecedented rates of biodiversity decline globally (Capdevila et al., 2022;Díaz et al., 2019). Our ability to reliably forecast ecosystems dynamics is limited by our capacity to understand what governs their composition, dynamics, function and structure (Dietze et al., 2018;Petchey et al., 2015). To drive predictive ecology forward and design appropriate conservation strategies, we therefore need access to long-term, highresolution and standardised information about ecosystems' abiotic and biotic components (Farley et al., 2018;

show abstract

Section: From Automated Data Collection To Ecological Knowledgementioning

confidence: 99%

Towards the fully automated monitoring of ecological communities

et al. 2022

View full text Add to dashboard Cite

show abstract

“…Following the findings of the LifeCLEF challenges (Joly et al, 2018 , 2019 , 2020 , 2021 ), AI-based identification of the world flora has improved significantly over the last 5 years, and it reached similar performance as human experts for common (Šulc et al, 2018 ) as well as for rare species (Picek et al, 2019 ). Ensembles of CNN models were able to recognize 10,000 plant species from Europe and North America and 10,000 from the Guiana shield and the Amazonia with approximately 90 and 40% accuracy, respectively.…”

Section: Related Workmentioning

confidence: 99%

“…For plant recognition, such large-scale data are available, thanks to citizen-science and open-data initiatives such as Encyclopedia of Life (EoL), Pl@ntNet, and the Global Biodiversity Information Facility (GBIF). This allowed building challenging datasets for fine-grained classification training and evaluation, e.g., in PlantCLEF (Goëau et al, 2016(Goëau et al, , 2017(Goëau et al, , 2018(Goëau et al, , 2020(Goëau et al, , 2021, LifeCLEF (Joly et al, 2018(Joly et al, , 2019(Joly et al, , 2020(Joly et al, , 2021, iNaturalist (Van Horn et al, 2018), and Pl@ntNet (Garcin et al, 2021).…”

Section: Introductionmentioning

confidence: 99%

Plant recognition by AI: Deep neural nets, transformers, and kNN in deep embeddings

et al. 2022

Self Cite

View full text Add to dashboard Cite

The article reviews and benchmarks machine learning methods for automatic image-based plant species recognition and proposes a novel retrieval-based method for recognition by nearest neighbor classification in a deep embedding space. The image retrieval method relies on a model trained via the Recall@k surrogate loss. State-of-the-art approaches to image classification, based on Convolutional Neural Networks (CNN) and Vision Transformers (ViT), are benchmarked and compared with the proposed image retrieval-based method. The impact of performance-enhancing techniques, e.g., class prior adaptation, image augmentations, learning rate scheduling, and loss functions, is studied. The evaluation is carried out on the PlantCLEF 2017, the ExpertLifeCLEF 2018, and the iNaturalist 2018 Datasets—the largest publicly available datasets for plant recognition. The evaluation of CNN and ViT classifiers shows a gradual improvement in classification accuracy. The current state-of-the-art Vision Transformer model, ViT-Large/16, achieves 91.15% and 83.54% accuracy on the PlantCLEF 2017 and ExpertLifeCLEF 2018 test sets, respectively; the best CNN model (ResNeSt-269e) error rate dropped by 22.91% and 28.34%. Apart from that, additional tricks increased the performance for the ViT-Base/32 by 3.72% on ExpertLifeCLEF 2018 and by 4.67% on PlantCLEF 2017. The retrieval approach achieved superior performance in all measured scenarios with accuracy margins of 0.28%, 4.13%, and 10.25% on ExpertLifeCLEF 2018, PlantCLEF 2017, and iNat2018–Plantae, respectively.

show abstract

“…We used a subset of the world's largest snake photo dataset, described in detail in Durso et al, 2021 [20], which we provided within the snake species identification challenge Snake-CLEF2021. This challenge is part of LifeCLEF21, the Conference and Labs of the Evaluation Forum (CLEF) that proposes data-oriented challenges related to the identification and prediction of biodiversity [23].…”

Section: Snake Photo Datasetsmentioning

confidence: 99%

An artificial intelligence model to identify snakes from across the world: Opportunities and challenges for global health and herpetology

Bolon

Picek²,

Durso³

et al. 2022

PLoS Negl Trop Dis

Self Cite

View full text Add to dashboard Cite

Background Snakebite envenoming is a neglected tropical disease that kills an estimated 81,000 to 138,000 people and disables another 400,000 globally every year. The World Health Organization aims to halve this burden by 2030. To achieve this ambitious goal, we need to close the data gap in snake ecology and snakebite epidemiology and give healthcare providers up-to-date knowledge and access to better diagnostic tools. An essential first step is to improve the capacity to identify biting snakes taxonomically. The existence of AI-based identification tools for other animals offers an innovative opportunity to apply machine learning to snake identification and snakebite envenoming, a life-threatening situation. Methodology We developed an AI model based on Vision Transformer, a recent neural network architecture, and a comprehensive snake photo dataset of 386,006 training photos covering 198 venomous and 574 non-venomous snake species from 188 countries. We gathered photos from online biodiversity platforms (iNaturalist and HerpMapper) and a photo-sharing site (Flickr). Principal findings The model macro-averaged F1 score, which reflects the species-wise performance as averaging performance for each species, is 92.2%. The accuracy on a species and genus level is 96.0% and 99.0%, respectively. The average accuracy per country is 94.2%. The model accurately classifies selected venomous and non-venomous lookalike species from Southeast Asia and sub-Saharan Africa. Conclusions To our knowledge, this model’s taxonomic and geographic coverage and performance are unprecedented. This model could provide high-speed and low-cost snake identification to support snakebite victims and healthcare providers in low-resource settings, as well as zoologists, conservationists, and nature lovers from across the world.

show abstract

Overview of LifeCLEF 2021: An Evaluation of Machine-Learning Based Species Identification and Species Distribution Prediction

Cited by 21 publications

References 40 publications

Towards the fully automated monitoring of ecological communities

Towards the fully automated monitoring of ecological communities

Plant recognition by AI: Deep neural nets, transformers, and kNN in deep embeddings

An artificial intelligence model to identify snakes from across the world: Opportunities and challenges for global health and herpetology

Contact Info

Product

Resources

About