Alexis Joly

Theo Larcher, Lukas Picek, Benjamin Deneu et al.

This paper describes a deep-SDM framework, MALPOLON. Written in Python and built upon the PyTorch library, this framework aims to facilitate training and inferences of deep species distribution models (deep-SDM) and sharing for users with only general Python language skills (e.g., modeling ecologists) who are interested in testing deep learning approaches to build new SDMs. More advanced users can also benefit from the framework's modularity to run more specific experiments by overriding existing classes while taking advantage of press-button examples to train neural networks on multiple classification tasks using custom or provided raw and pre-processed datasets. The framework is open-sourced on GitHub and PyPi along with extensive documentation and examples of use in various scenarios. MALPOLON offers straightforward installation, YAML-based configuration, parallel computing, multi-GPU utilization, baseline and foundational models for benchmarking, and extensive tutorials/documentation, aiming to enhance accessibility and performance scalability for ecologists and researchers.

Diego Marcos, Robert van de Vlasakker, Ioannis N. Athanasiadis et al.

Plant morphological traits, their observable characteristics, are fundamental to understand the role played by each species within their ecosystem. However, compiling trait information for even a moderate number of species is a demanding task that may take experts years to accomplish. At the same time, massive amounts of information about species descriptions is available online in the form of text, although the lack of structure makes this source of data impossible to use at scale. To overcome this, we propose to leverage recent advances in large language models (LLMs) and devise a mechanism for gathering and processing information on plant traits in the form of unstructured textual descriptions, without manual curation. We evaluate our approach by automatically replicating three manually created species-trait matrices. Our method managed to find values for over half of all species-trait pairs, with an F1-score of over 75%. Our results suggest that large-scale creation of structured trait databases from unstructured online text is currently feasible thanks to the information extraction capabilities of LLMs, being limited by the availability of textual descriptions covering all the traits of interest.

Tanguy Lefort, Benjamin Charlier, Alexis Joly et al.

In supervised learning - for instance in image classification - modern massive datasets are commonly labeled by a crowd of workers. The obtained labels in this crowdsourcing setting are then aggregated for training, generally leveraging a per-worker trust score. Yet, such workers oriented approaches discard the tasks' ambiguity. Ambiguous tasks might fool expert workers, which is often harmful for the learning step. In standard supervised learning settings - with one label per task - the Area Under the Margin (AUM) was tailored to identify mislabeled data. We adapt the AUM to identify ambiguous tasks in crowdsourced learning scenarios, introducing the Weighted Areas Under the Margin (WAUM). The WAUM is an average of AUMs weighted according to task-dependent scores. We show that the WAUM can help discarding ambiguous tasks from the training set, leading to better generalization performance. We report improvements over existing strategies for learning with a crowd, both on simulated settings, and on real datasets such as CIFAR-10H (a crowdsourced dataset with a high number of answered labels),LabelMe and Music (two datasets with few answered votes).

Lukas Picek, Christophe Botella, Maximilien Servajean et al.

The difficulty of monitoring biodiversity at fine scales and over large areas limits ecological knowledge and conservation efforts. To fill this gap, Species Distribution Models (SDMs) predict species across space from spatially explicit features. Yet, they face the challenge of integrating the rich but heterogeneous data made available over the past decade, notably millions of opportunistic species observations and standardized surveys, as well as multimodal remote sensing data. In light of that, we have designed and developed a new European-scale dataset for SDMs at high spatial resolution (10--50m), including more than 10k species (i.e., most of the European flora). The dataset comprises 5M heterogeneous Presence-Only records and 90k exhaustive Presence-Absence survey records, all accompanied by diverse environmental rasters (e.g., elevation, human footprint, and soil) traditionally used in SDMs. In addition, it provides Sentinel-2 RGB and NIR satellite images with 10 m resolution, a 20-year time series of climatic variables, and satellite time series from the Landsat program. In addition to the data, we provide an openly accessible SDM benchmark (hosted on Kaggle), which has already attracted an active community and a set of strong baselines for single predictor/modality and multimodal approaches. All resources, e.g., the dataset, pre-trained models, and baseline methods (in the form of notebooks), are available on Kaggle, allowing one to start with our dataset literally with two mouse clicks.

Ilyass Moummad, Marius Miron, Lukas Rauch et al.

Audio-to-image retrieval offers an interpretable alternative to audio-only classification for bioacoustic species recognition, but learning aligned audio-image representations is challenging due to the scarcity of paired audio-image data. We propose a simple and data-efficient approach that enables audio-to-image retrieval without any audio-image supervision. Our proposed method uses text as a semantic intermediary: we distill the text embedding space of a pretrained image-text model (BioCLIP-2), which encodes rich visual and taxonomic structure, into a pretrained audio-text model (BioLingual) by fine-tuning its audio encoder with a contrastive objective. This distillation transfers visually grounded semantics into the audio representation, inducing emergent alignment between audio and image embeddings without using images during training. We evaluate the resulting model on multiple bioacoustic benchmarks. The distilled audio encoder preserves audio discriminative power while substantially improving audio-text alignment on focal recordings and soundscape datasets. Most importantly, on the SSW60 benchmark, the proposed approach achieves strong audio-to-image retrieval performance exceeding baselines based on zero-shot model combinations or learned mappings between text embeddings, despite not training on paired audio-image data. These results demonstrate that indirect semantic transfer through text is sufficient to induce meaningful audio-image alignment, providing a practical solution for visually grounded species recognition in data-scarce bioacoustic settings.

Ilyass Moummad, Marius Miron, David Robinson et al.

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.

Camille Garcin, Maximilien Servajean, Alexis Joly et al.

Average-K classification is an alternative to top-K classification in which the number of labels returned varies with the ambiguity of the input image but must average to K over all the samples. A simple method to solve this task is to threshold the softmax output of a model trained with the cross-entropy loss. This approach is theoretically proven to be asymptotically consistent, but it is not guaranteed to be optimal for a finite set of samples. In this paper, we propose a new loss function based on a multi-label classification head in addition to the classical softmax. This second head is trained using pseudo-labels generated by thresholding the softmax head while guaranteeing that K classes are returned on average. We show that this approach allows the model to better capture ambiguities between classes and, as a result, to return more consistent sets of possible classes. Experiments on two datasets from the literature demonstrate that our approach outperforms the softmax baseline, as well as several other loss functions more generally designed for weakly supervised multi-label classification. The gains are larger the higher the uncertainty, especially for classes with few samples.

Quentin Leroy, Olivier Buisson, Alexis Joly

Supervised representation learning with deep networks tends to overfit the training classes and the generalization to novel classes is a challenging question. It is common to evaluate a learned embedding on held-out images of the same training classes. In real applications however, data comes from new sources and novel classes are likely to arise. We hypothesize that incorporating unlabelled images of novel classes in the training set in a semi-supervised fashion would be beneficial for the efficient retrieval of novel-class images compared to a vanilla supervised representation. To verify this hypothesis in a comprehensive way, we propose an original evaluation methodology that varies the degree of novelty of novel classes by partitioning the dataset category-wise either randomly, or semantically, i.e. by minimizing the shared semantics between base and novel classes. This evaluation procedure allows to train a representation blindly to any novel-class labels and evaluate the frozen representation on the retrieval of base or novel classes. We find that a vanilla supervised representation falls short on the retrieval of novel classes even more so when the semantics gap is higher. Semi-supervised algorithms allow to partially bridge this performance gap but there is still much room for improvement.

Kawtar Zaher, Olivier Buisson, Alexis Joly

Real-world fine-grained visual retrieval often requires discovering a rare concept from large unlabeled collections with minimal supervision. This is especially critical in biodiversity monitoring, ecological studies, and long-tailed visual domains, where the target may represent only a tiny fraction of the data, creating highly imbalanced binary problems. Interactive retrieval with relevance feedback offers a practical solution: starting from a small query, the system selects candidates for binary user annotation and iteratively refines a lightweight classifier. While Active Learning (AL) is commonly used to guide selection, conventional AL assumes symmetric class priors and large annotation budgets, limiting effectiveness in imbalanced, low-budget, low-latency settings. We introduce Positive-First Most Ambiguous (PF-MA), a simple yet effective AL criterion that explicitly addresses the class imbalance asymmetry: it prioritizes near-boundary samples while favoring likely positives, enabling rapid discovery of subtle visual categories while maintaining informativeness. Unlike standard methods that oversample negatives, PF-MA consistently returns small batches with a high proportion of relevant samples, improving early retrieval and user satisfaction. To capture retrieval diversity, we also propose a class coverage metric that measures how well selected positives span the visual variability of the target class. Experiments on long-tailed datasets, including fine-grained botanical data, demonstrate that PF-MA consistently outperforms strong baselines in both coverage and classifier performance, across varying class sizes and descriptors. Our results highlight that aligning AL with the asymmetric and user-centric objectives of interactive fine-grained retrieval enables simple yet powerful solutions for retrieving rare and visually subtle categories in realistic human-in-the-loop settings.

Sébastien Gigot--Léandri, Gaétan Morand, Alexis Joly et al.

Species distribution models (SDMs) commonly produce probabilistic occurrence predictions that must be converted into binary presence-absence maps for ecological inference and conservation planning. However, this binarization step is typically heuristic and can substantially distort estimates of species prevalence and community composition. We present MaxExp, a decision-driven binarization framework that selects the most probable species assemblage by directly maximizing a chosen evaluation metric. MaxExp requires no calibration data and is flexible across several scores. We also introduce the Set Size Expectation (SSE) method, a computationally efficient alternative that predicts assemblages based on expected species richness. Using three case studies spanning diverse taxa, species counts, and performance metrics, we show that MaxExp consistently matches or surpasses widely used thresholding and calibration methods, especially under strong class imbalance and high rarity. SSE offers a simpler yet competitive option. Together, these methods provide robust, reproducible tools for multispecies SDM binarization.

Kawtar Zaher, Ilyass Moummad, Olivier Buisson et al.

Most self-supervised learning (SSL) methods learn continuous visual representations by aligning different views of the same input, offering limited control over how information is structured across representation dimensions. In this work, we frame visual self-supervised learning as a discrete communication process between a teacher and a student network, where semantic information is transmitted through a fixed-capacity binary channel. Rather than aligning continuous features, the student predicts multi-label binary messages produced by the teacher. Discrete agreement is enforced through an element-wise binary cross-entropy objective, while a coding-rate regularization term encourages effective utilization of the constrained channel, promoting structured representations. We further show that periodically reinitializing the projection head strengthens this effect by encouraging embeddings that remain predictive across multiple discrete encodings. Extensive experiments demonstrate consistent improvements over continuous agreement baselines on image classification, retrieval, and dense visual prediction tasks, as well as under domain shift through self-supervised adaptation. Beyond backbone representations, we analyze the learned binary codes and show that they form a compact and informative discrete language, capturing semantic factors reusable across classes.

Ilyass Moummad, Kawtar Zaher, Hervé Goëau et al.

Efficient large-scale retrieval requires representations that are both compact and discriminative. Foundation models provide powerful visual and multimodal embeddings, but nearest neighbor search in these high-dimensional spaces is computationally expensive. Hashing offers an efficient alternative by enabling fast Hamming distance search with binary codes, yet existing approaches often rely on complex pipelines, multi-term objectives, designs specialized for a single learning paradigm, and long training times. We introduce CroVCA (Cross-View Code Alignment), a simple and unified principle for learning binary codes that remain consistent across semantically aligned views. A single binary cross-entropy loss enforces alignment, while coding-rate maximization serves as an anti-collapse regularizer to promote balanced and diverse codes. To implement this, we design HashCoder, a lightweight MLP hashing network with a final batch normalization layer to enforce balanced codes. HashCoder can be used as a probing head on frozen embeddings or to adapt encoders efficiently via LoRA fine-tuning. Across benchmarks, CroVCA achieves state-of-the-art results in just 5 training epochs. At 16 bits, it particularly well-for instance, unsupervised hashing on COCO completes in under 2 minutes and supervised hashing on ImageNet100 in about 3 minutes on a single GPU. These results highlight CroVCA's efficiency, adaptability, and broad applicability.

Herve Goeau, Pierre Bonnet, Alexis Joly

The 2017-th edition of the LifeCLEF plant identification challenge is an important milestone towards automated plant identification systems working at the scale of continental floras with 10.000 plant species living mainly in Europe and North America illustrated by a total of 1.1M images. Nowadays, such ambitious systems are enabled thanks to the conjunction of the dazzling recent progress in image classification with deep learning and several outstanding international initiatives, such as the Encyclopedia of Life (EOL), aggregating the visual knowledge on plant species coming from the main national botany institutes. However, despite all these efforts the majority of the plant species still remain without pictures or are poorly illustrated. Outside the institutional channels, a much larger number of plant pictures are available and spread on the web through botanist blogs, plant lovers web-pages, image hosting websites and on-line plant retailers. The LifeCLEF 2017 plant challenge presented in this paper aimed at evaluating to what extent a large noisy training dataset collected through the web and containing a lot of labelling errors can compete with a smaller but trusted training dataset checked by experts. To fairly compare both training strategies, the test dataset was created from a third data source, i.e. the Pl@ntNet mobile application that collects millions of plant image queries all over the world. This paper presents more precisely the resources and assessments of the challenge, summarizes the approaches and systems employed by the participating research groups, and provides an analysis of the main outcomes.

Herve Goeau, Pierre Bonnet, Alexis Joly

The LifeCLEF plant identification challenge aims at evaluating plant identification methods and systems at a very large scale, close to the conditions of a real-world biodiversity monitoring scenario. The 2016-th edition was actually conducted on a set of more than 110K images illustrating 1000 plant species living in West Europe, built through a large-scale participatory sensing platform initiated in 2011 and which now involves tens of thousands of contributors. The main novelty over the previous years is that the identification task was evaluated as an open-set recognition problem, i.e. a problem in which the recognition system has to be robust to unknown and never seen categories. Beyond the brute-force classification across the known classes of the training set, the big challenge was thus to automatically reject the false positive classification hits that are caused by the unknown classes. This overview presents more precisely the resources and assessments of the challenge, summarizes the approaches and systems employed by the participating research groups, and provides an analysis of the main outcomes.

Herve Goeau, Alexis Joly, Pierre Bonnet et al.

The LifeCLEFs plant identification task provides a testbed for a system-oriented evaluation of plant identification about 500 species trees and herbaceous plants. Seven types of image content are considered: scan and scan-like pictures of leaf, and 6 kinds of detailed views with unconstrained conditions, directly photographed on the plant: flower, fruit, stem & bark, branch, leaf and entire view. The main originality of this data is that it was specifically built through a citizen sciences initiative conducted by Tela Botanica, a French social network of amateur and expert botanists. This makes the task closer to the conditions of a real-world application. This overview presents more precisely the resources and assessments of task, summarizes the retrieval approaches employed by the participating groups, and provides an analysis of the main evaluation results. With a total of ten groups from six countries and with a total of twenty seven submitted runs, involving distinct and original methods, this fourth year task confirms Image & Multimedia Retrieval community interest for biodiversity and botany, and highlights further challenging studies in plant identification.

Herve Goeau, Pierre Bonnet, Alexis Joly

The LifeCLEF plant identification challenge aims at evaluating plant identification methods and systems at a very large scale, close to the conditions of a real-world biodiversity monitoring scenario. The 2015 evaluation was actually conducted on a set of more than 100K images illustrating 1000 plant species living in West Europe. The main originality of this dataset is that it was built through a large-scale participatory sensing plateform initiated in 2011 and which now involves tens of thousands of contributors. This overview presents more precisely the resources and assessments of the challenge, summarizes the approaches and systems employed by the participating research groups, and provides an analysis of the main outcomes.

Herve Goeau, Pierre Bonnet, Alexis Joly

Automated identification of plants has improved considerably thanks to the recent progress in deep learning and the availability of training data. However, this profusion of data only concerns a few tens of thousands of species, while the planet has nearly 369K. The LifeCLEF 2019 Plant Identification challenge (or "PlantCLEF 2019") was designed to evaluate automated identification on the flora of data deficient regions. It is based on a dataset of 10K species mainly focused on the Guiana shield and the Northern Amazon rainforest, an area known to have one of the greatest diversity of plants and animals in the world. As in the previous edition, a comparison of the performance of the systems evaluated with the best tropical flora experts was carried out. This paper presents the resources and assessments of the challenge, summarizes the approaches and systems employed by the participating research groups, and provides an analysis of the main outcomes.

Herve Goeau, Pierre Bonnet, Alexis Joly

It is estimated that there are more than 300,000 species of vascular plants in the world. Increasing our knowledge of these species is of paramount importance for the development of human civilization (agriculture, construction, pharmacopoeia, etc.), especially in the context of the biodiversity crisis. However, the burden of systematic plant identification by human experts strongly penalizes the aggregation of new data and knowledge. Since then, automatic identification has made considerable progress in recent years as highlighted during all previous editions of PlantCLEF. Deep learning techniques now seem mature enough to address the ultimate but realistic problem of global identification of plant biodiversity in spite of many problems that the data may present (a huge number of classes, very strongly unbalanced classes, partially erroneous identifications, duplications, variable visual quality, diversity of visual contents such as photos or herbarium sheets, etc). The PlantCLEF2022 challenge edition proposes to take a step in this direction by tackling a multi-image (and metadata) classification problem with a very large number of classes (80k plant species). This paper presents the resources and evaluations of the challenge, summarizes the approaches and systems employed by the participating research groups, and provides an analysis of key findings.

Herve Goeau, Pierre Bonnet, Alexis Joly

Automated identification of plants and animals has improved considerably in the last few years, in particular thanks to the recent advances in deep learning. The next big question is how far such automated systems are from the human expertise. Indeed, even the best experts are sometimes confused and/or disagree between each others when validating visual or audio observations of living organism. A picture actually contains only a partial information that is usually not sufficient to determine the right species with certainty. Quantifying this uncertainty and comparing it to the performance of automated systems is of high interest for both computer scientists and expert naturalists. The LifeCLEF 2018 ExpertCLEF challenge presented in this paper was designed to allow this comparison between human experts and automated systems. In total, 19 deep-learning systems implemented by 4 different research teams were evaluated with regard to 9 expert botanists of the French flora. The main outcome of this work is that the performance of state-of-the-art deep learning models is now close to the most advanced human expertise. This paper presents more precisely the resources and assessments of the challenge, summarizes the approaches and systems employed by the participating research groups, and provides an analysis of the main outcomes.

Herve Goeau, Pierre Bonnet, Alexis Joly

Automated plant identification has improved considerably thanks to recent advances in deep learning and the availability of training data with more and more field photos. However, this profusion of data concerns only a few tens of thousands of species, mainly located in North America and Western Europe, much less in the richest regions in terms of biodiversity such as tropical countries. On the other hand, for several centuries, botanists have systematically collected, catalogued and stored plant specimens in herbaria, especially in tropical regions, and recent efforts by the biodiversity informatics community have made it possible to put millions of digitised records online. The LifeCLEF 2021 plant identification challenge (or "PlantCLEF 2021") was designed to assess the extent to which automated identification of flora in data-poor regions can be improved by using herbarium collections. It is based on a dataset of about 1,000 species mainly focused on the Guiana Shield of South America, a region known to have one of the highest plant diversities in the world. The challenge was evaluated as a cross-domain classification task where the training set consisted of several hundred thousand herbarium sheets and a few thousand photos to allow learning a correspondence between the two domains. In addition to the usual metadata (location, date, author, taxonomy), the training data also includes the values of 5 morphological and functional traits for each species. The test set consisted exclusively of photos taken in the field. This article presents the resources and evaluations of the assessment carried out, summarises the approaches and systems used by the participating research groups and provides an analysis of the main results.

Herve Goeau, Pierre Bonnet, Alexis Joly

Automated identification of plants has improved considerably thanks to the recent progress in deep learning and the availability of training data with more and more photos in the field. However, this profusion of data only concerns a few tens of thousands of species, mostly located in North America and Western Europe, much less in the richest regions in terms of biodiversity such as tropical countries. On the other hand, for several centuries, botanists have collected, catalogued and systematically stored plant specimens in herbaria, particularly in tropical regions, and the recent efforts by the biodiversity informatics community made it possible to put millions of digitized sheets online. The LifeCLEF 2020 Plant Identification challenge (or "PlantCLEF 2020") was designed to evaluate to what extent automated identification on the flora of data deficient regions can be improved by the use of herbarium collections. It is based on a dataset of about 1,000 species mainly focused on the South America's Guiana Shield, an area known to have one of the greatest diversity of plants in the world. The challenge was evaluated as a cross-domain classification task where the training set consist of several hundred thousand herbarium sheets and few thousand of photos to enable learning a mapping between the two domains. The test set was exclusively composed of photos in the field. This paper presents the resources and assessments of the conducted evaluation, summarizes the approaches and systems employed by the participating research groups, and provides an analysis of the main outcomes.

Christophe Botella, Benjamin Deneu, Diego Marcos et al.

Understanding the spatio-temporal distribution of species is a cornerstone of ecology and conservation. By pairing species observations with geographic and environmental predictors, researchers can model the relationship between an environment and the species which may be found there. To advance the state-of-the-art in this area with deep learning models and remote sensing data, we organized an open machine learning challenge called GeoLifeCLEF 2023. The training dataset comprised 5 million plant species observations (single positive label per sample) distributed across Europe and covering most of its flora, high-resolution rasters: remote sensing imagery, land cover, elevation, in addition to coarse-resolution data: climate, soil and human footprint variables. In this multi-label classification task, we evaluated models ability to predict the species composition in 22 thousand small plots based on standardized surveys. This paper presents an overview of the competition, synthesizes the approaches used by the participating teams, and analyzes the main results. In particular, we highlight the biases faced by the methods fitted to single positive labels when it comes to the multi-label evaluation, and the new and effective learning strategy combining single and multi-label data in training.

Ilyass Moummad, Kawtar Zaher, Hervé Goëau et al.

Automated plant recognition plays a crucial role in biodiversity monitoring and conservation, yet current approaches rely heavily on supervised learning, which is limited by the availability of expert-labeled data. Self-supervised learning (SSL) offers a scalable alternative, but existing methods and training protocols are largely designed for coarse-grained visual tasks and may not transfer well to fine-grained domains such as plant species recognition. In this work, we investigate SSL for plant image representation learning. We show that commonly used augmentations in SSL pipelines - such as Gaussian blur, grayscale conversion, and solarization - are detrimental in the context of plant images, as they remove subtle discriminative cues essential for fine-grained recognition. We instead identify alternative transformations, including affine and posterization, that are better suited to this domain. We further demonstrate that training SimDINOv2 on the iNaturalist 2021 Plantae subset yields significantly stronger representations than training on ImageNet-1K, highlighting the importance of domain-specific data for SSL. Our findings are consistent across both ViT-Base and ViT-Large architectures. Moreover, our models achieve competitive performance and sometimes outperform strong supervised baselines Pl@ntCLEF and BioCLIP on downstream plant recognition tasks in few-shot settings. Overall, our results highlight the critical importance of domain-adapted augmentation strategies and dataset selection in self-supervised learning, and provide practical guidelines for building scalable models for biodiversity monitoring.

Herve Goeau, Pierre Bonnet, Alexis Joly

The world is estimated to be home to over 300,000 species of vascular plants. In the face of the ongoing biodiversity crisis, expanding our understanding of these species is crucial for the advancement of human civilization, encompassing areas such as agriculture, construction, and pharmacopoeia. However, the labor-intensive process of plant identification undertaken by human experts poses a significant obstacle to the accumulation of new data and knowledge. Fortunately, recent advancements in automatic identification, particularly through the application of deep learning techniques, have shown promising progress. Despite challenges posed by data-related issues such as a vast number of classes, imbalanced class distribution, erroneous identifications, duplications, variable visual quality, and diverse visual contents (such as photos or herbarium sheets), deep learning approaches have reached a level of maturity which gives us hope that in the near future we will have an identification system capable of accurately identifying all plant species worldwide. The PlantCLEF2023 challenge aims to contribute to this pursuit by addressing a multi-image (and metadata) classification problem involving an extensive set of classes (80,000 plant species). This paper provides an overview of the challenge's resources and evaluations, summarizes the methods and systems employed by participating research groups, and presents an analysis of key findings.

Ilyass Moummad, Reda Bensaid, Kawtar Zaher et al.

Recent advances in large-scale visual representation learning have significantly improved performance in plant species and plant disease recognition tasks. However, state-of-the-art models, often based on high-capacity vision transformers or multimodal foundation models, remain computationally expensive and difficult to deploy in resource-constrained environments such as mobile or edge devices. This limitation hinders the scalability of automated biodiversity monitoring and precision agriculture systems, where efficiency is as critical as accuracy. In this work, we investigate knowledge distillation as an effective approach to transfer the representational capacity of large pretrained models into smaller, more efficient architectures. We focus on plant species and disease recognition, and conduct an extensive empirical study on two challenging benchmarks: Pl@ntNet300K-v2 and Deep-Plant-Disease. We evaluate four representative architectures, including two ConvNeXt models and two vision transformers, under multiple training regimes: from-scratch training and pretrained initialization, each with and without distillation. In total, we train and evaluate 70 models. Our results show that knowledge distillation consistently improves performance across tasks and architectures. Distilled models are able to match the performance of significantly larger models while maintaining substantially lower computational cost. These findings demonstrate the potential of knowledge distillation techniques to enable efficient and scalable deployment of plant recognition systems in real-world environmental applications.

Herve Goeau, Vincent Espitalier, Pierre Bonnet et al.

Plot images are essential for ecological studies, enabling standardized sampling, biodiversity assessment, long-term monitoring and remote, large-scale surveys. Plot images are typically fifty centimetres or one square meter in size, and botanists meticulously identify all the species found there. The integration of AI could significantly improve the efficiency of specialists, helping them to extend the scope and coverage of ecological studies. To evaluate advances in this regard, the PlantCLEF 2024 challenge leverages a new test set of thousands of multi-label images annotated by experts and covering over 800 species. In addition, it provides a large training set of 1.7 million individual plant images as well as state-of-the-art vision transformer models pre-trained on this data. The task is evaluated as a (weakly-labeled) multi-label classification task where the aim is to predict all the plant species present on a high-resolution plot image (using the single-label training data). In this paper, we provide an detailed description of the data, the evaluation methodology, the methods and models employed by the participants and the results achieved.

Lukas Rauch, René Heinrich, Ilyass Moummad et al.

Masked Autoencoders (MAEs) learn rich semantic representations in audio classification through an efficient self-supervised reconstruction task. However, general-purpose models fail to generalize well when applied directly to fine-grained audio domains. Specifically, bird-sound classification requires distinguishing subtle inter-species differences and managing high intra-species acoustic variability, revealing the performance limitations of general-domain Audio-MAEs. This work demonstrates that bridging this domain gap domain gap requires full-pipeline adaptation, not just domain-specific pretraining data. We systematically revisit and adapt the pretraining recipe, fine-tuning methods, and frozen feature utilization to bird sounds using BirdSet, a large-scale bioacoustic dataset comparable to AudioSet. Our resulting Bird-MAE achieves new state-of-the-art results in BirdSet's multi-label classification benchmark. Additionally, we introduce the parameter-efficient prototypical probing, enhancing the utility of frozen MAE representations and closely approaching fine-tuning performance in low-resource settings. Bird-MAE's prototypical probes outperform linear probing by up to 37 percentage points in mean average precision and narrow the gap to fine-tuning across BirdSet downstream tasks. Bird-MAE also demonstrates robust few-shot capabilities with prototypical probing in our newly established few-shot benchmark on BirdSet, highlighting the potential of tailored self-supervised learning pipelines for fine-grained audio domains.

Giulio Martellucci, Herve Goeau, Pierre Bonnet et al.

Quadrat images are essential for ecological studies, as they enable standardized sampling, the assessment of plant biodiversity, long-term monitoring, and large-scale field campaigns. These images typically cover an area of fifty centimetres or one square meter, and botanists carefully identify all the species present. Integrating AI could help specialists accelerate their inventories and expand the spatial coverage of ecological studies. To assess progress in this area, the PlantCLEF 2025 challenge relies on a new test set of 2,105 high-resolution multi-label images annotated by experts and covering around 400 species. It also provides a large training set of 1.4 million individual plant images, along with vision transformer models pre-trained on this data. The task is formulated as a (weakly labelled) multi-label classification problem, where the goal is to predict all species present in a quadrat image using single-label training data. This paper provides a detailed description of the data, the evaluation methodology, the methods and models used by participants, and the results achieved.

Joaquim Estopinan, Pierre Bonnet, Maximilien Servajean et al.

The post-2020 global biodiversity framework needs ambitious, research-based targets. Estimating the accelerated extinction risk due to climate change is critical. The International Union for Conservation of Nature (IUCN) measures the extinction risk of species. Automatic methods have been developed to provide information on the IUCN status of under-assessed taxa. However, these compensatory methods are based on current species characteristics, mainly geographical, which precludes their use in future projections. Here, we evaluate a novel method for classifying the IUCN status of species benefiting from the generalisation power of species distribution models based on deep learning. Our method matches state-of-the-art classification performance while relying on flexible SDM-based features that capture species' environmental preferences. Cross-validation yields average accuracies of 0.61 for status classification and 0.78 for binary classification. Climate change will reshape future species distributions. Under the species-environment equilibrium hypothesis, SDM projections approximate plausible future outcomes. Two extremes of species dispersal capacity are considered: unlimited or null. The projected species distributions are translated into features feeding our IUCN classification method. Finally, trends in threatened species are analysed over time and i) by continent and as a function of average ii) latitude or iii) altitude. The proportion of threatened species is increasing globally, with critical rates in Africa, Asia and South America. Furthermore, the proportion of threatened species is predicted to peak around the two Tropics, at the Equator, in the lowlands and at altitudes of 800-1,500 m.

Joaquim Estopinan, Maximilien Servajean, Pierre Bonnet et al.

Although increasing threats on biodiversity are now widely recognised, there are no accurate global maps showing whether and where species assemblages are at risk. We hereby assess and map at kilometre resolution the conservation status of the iconic orchid family, and discuss the insights conveyed at multiple scales. We introduce a new Deep Species Distribution Model trained on 1M occurrences of 14K orchid species to predict their assemblages at global scale and at kilometre resolution. We propose two main indicators of the conservation status of the assemblages: (i) the proportion of threatened species, and (ii) the status of the most threatened species in the assemblage. We show and analyze the variation of these indicators at World scale and in relation to currently protected areas in Sumatra island. Global and interactive maps available online show the indicators of conservation status of orchid assemblages, with sharp spatial variations at all scales. The highest level of threat is found at Madagascar and the neighbouring islands. In Sumatra, we found good correspondence of protected areas with our indicators, but supplementing current IUCN assessments with status predictions results in alarming levels of species threat across the island. Recent advances in deep learning enable reliable mapping of the conservation status of species assemblages on a global scale. As an umbrella taxon, orchid family provides a reference for identifying vulnerable ecosystems worldwide, and prioritising conservation actions both at international and local levels.

César Leblanc, Lukas Picek, Benjamin Deneu et al.

This paper describes a cascading multimodal pipeline for high-resolution biodiversity mapping across Europe, integrating species distribution modeling, biodiversity indicators, and habitat classification. The proposed pipeline first predicts species compositions using a deep-SDM, a multimodal model trained on remote sensing, climate time series, and species occurrence data at 50x50m resolution. These predictions are then used to generate biodiversity indicator maps and classify habitats with Pl@ntBERT, a transformer-based LLM designed for species-to-habitat mapping. With this approach, continental-scale species distribution maps, biodiversity indicator maps, and habitat maps are produced, providing fine-grained ecological insights. Unlike traditional methods, this framework enables joint modeling of interspecies dependencies, bias-aware training with heterogeneous presence-absence data, and large-scale inference from multi-source remote sensing inputs.

Ilyass Moummad, Kawtar Zaher, Lukas Rauch et al.

Information retrieval with compact binary embeddings, also referred to as hashing, is crucial for scalable fast search applications, yet state-of-the-art hashing methods require expensive, scenario-specific training. In this work, we introduce Hashing-Baseline, a strong training-free hashing method leveraging powerful pretrained encoders that produce rich pretrained embeddings. We revisit classical, training-free hashing techniques: principal component analysis, random orthogonal projection, and threshold binarization, to produce a strong baseline for hashing. Our approach combines these techniques with frozen embeddings from state-of-the-art vision and audio encoders to yield competitive retrieval performance without any additional learning or fine-tuning. To demonstrate the generality and effectiveness of this approach, we evaluate it on standard image retrieval benchmarks as well as a newly introduced benchmark for audio hashing.

Kawtar Zaher, Olivier Buisson, Alexis Joly

Building on existing approaches, we revisit Human-in-the-Loop Object Retrieval, a task that consists of iteratively retrieving images containing objects of a class-of-interest, specified by a user-provided query. Starting from a large unlabeled image collection, the aim is to rapidly identify diverse instances of an object category relying solely on the initial query and the user's Relevance Feedback, with no prior labels. The retrieval process is formulated as a binary classification task, where the system continuously learns to distinguish between relevant and non-relevant images to the query, through iterative user interaction. This interaction is guided by an Active Learning loop: at each iteration, the system selects informative samples for user annotation, thereby refining the retrieval performance. This task is particularly challenging in multi-object datasets, where the object of interest may occupy only a small region of the image within a complex, cluttered scene. Unlike object-centered settings where global descriptors often suffice, multi-object images require more adapted, localized descriptors. In this work, we formulate and revisit the Human-in-the-Loop Object Retrieval task by leveraging pre-trained ViT representations, and addressing key design questions, including which object instances to consider in an image, what form the annotations should take, how Active Selection should be applied, and which representation strategies best capture the object's features. We compare several representation strategies across multi-object datasets highlighting trade-offs between capturing the global context and focusing on fine-grained local object details. Our results offer practical insights for the design of effective interactive retrieval pipelines based on Active Learning for object class retrieval.

Lukas Picek, César Leblanc, Alexis Joly et al.

This paper describes GeoPl@ntNet, an interactive web application designed to make Essential Biodiversity Variables accessible and understandable to everyone through dynamic maps and fact sheets. Its core purpose is to allow users to explore high-resolution AI-generated maps of species distributions, habitat types, and biodiversity indicators across Europe. These maps, developed through a cascading pipeline involving convolutional neural networks and large language models, provide an intuitive yet information-rich interface to better understand biodiversity, with resolutions as precise as 50x50 meters. The website also enables exploration of specific regions, allowing users to select areas of interest on the map (e.g., urban green spaces, protected areas, or riverbanks) to view local species and their coverage. Additionally, GeoPl@ntNet generates comprehensive reports for selected regions, including insights into the number of protected species, invasive species, and endemic species.

Simon Ravé, Jean-Christophe Lombardo, Pejman Rasti et al.

We present a zero-shot segmentation approach for agricultural imagery that leverages Plantnet, a large-scale plant classification model, in conjunction with its DinoV2 backbone and the Segment Anything Model (SAM). Rather than collecting and annotating new datasets, our method exploits Plantnet's specialized plant representations to identify plant regions and produce coarse segmentation masks. These masks are then refined by SAM to yield detailed segmentations. We evaluate on four publicly available datasets of various complexity in terms of contrast including some where the limited size of the training data and complex field conditions often hinder purely supervised methods. Our results show consistent performance gains when using Plantnet-fine-tuned DinoV2 over the base DinoV2 model, as measured by the Jaccard Index (IoU). These findings highlight the potential of combining foundation models with specialized plant-centric models to alleviate the annotation bottleneck and enable effective segmentation in diverse agricultural scenarios.

Matteo Contini, Victor Illien, Sylvain Poulain et al.

Monitoring coral reefs at large spatial scales remains an open challenge, essential for assessing ecosystem health and informing conservation efforts. While drone-based aerial imagery offers broad spatial coverage, its limited resolution makes it difficult to reliably distinguish fine-scale classes, such as coral morphotypes. At the same time, obtaining pixel-level annotations over large spatial extents is costly and labor-intensive, limiting the scalability of deep learning-based segmentation methods for aerial imagery. We present a multi-scale weakly supervised semantic segmentation framework that addresses this challenge by transferring fine-scale ecological information from underwater imagery to aerial data. Our method enables large-scale coral reef mapping from drone imagery with minimal manual annotation, combining classification-based supervision, spatial interpolation and self-distillation techniques. We demonstrate the efficacy of the approach, enabling large-area segmentation of coral morphotypes and demonstrating flexibility for integrating new classes. This study presents a scalable, cost-effective methodology for high-resolution reef monitoring, combining low-cost data collection, weakly supervised deep learning and multi-scale remote sensing.

Matteo Contini, Victor Illien, Julien Barde et al.

Drone-based remote sensing combined with AI-driven methodologies has shown great potential for accurate mapping and monitoring of coral reef ecosystems. This study presents a novel multi-scale approach to coral reef monitoring, integrating fine-scale underwater imagery with medium-scale aerial imagery. Underwater images are captured using an Autonomous Surface Vehicle (ASV), while aerial images are acquired with an aerial drone. A transformer-based deep-learning model is trained on underwater images to detect the presence of 31 classes covering various coral morphotypes, associated fauna, and habitats. These predictions serve as annotations for training a second model applied to aerial images. The transfer of information across scales is achieved through a weighted footprint method that accounts for partial overlaps between underwater image footprints and aerial image tiles. The results show that the multi-scale methodology successfully extends fine-scale classification to larger reef areas, achieving a high degree of accuracy in predicting coral morphotypes and associated habitats. The method showed a strong alignment between underwater-derived annotations and ground truth data, reflected by an AUC (Area Under the Curve) score of 0.9251. This shows that the integration of underwater and aerial imagery, supported by deep-learning models, can facilitate scalable and accurate reef assessments. This study demonstrates the potential of combining multi-scale imaging and AI to facilitate the monitoring and conservation of coral reefs. Our approach leverages the strengths of underwater and aerial imagery, ensuring the precision of fine-scale analysis while extending it to cover a broader reef area.

Maxime Ryckewaert, Diego Marcos, Christophe Botella et al.

The rapid expansion of citizen science initiatives has led to a significant growth of biodiversity databases, and particularly presence-only (PO) observations. PO data are invaluable for understanding species distributions and their dynamics, but their use in a Species Distribution Model (SDM) is curtailed by sampling biases and the lack of information on absences. Poisson point processes are widely used for SDMs, with Maxent being one of the most popular methods. Maxent maximises the entropy of a probability distribution across sites as a function of predefined transformations of variables, called features. In contrast, neural networks and deep learning have emerged as a promising technique for automatic feature extraction from complex input variables. Arbitrarily complex transformations of input variables can be learned from the data efficiently through backpropagation and stochastic gradient descent (SGD). In this paper, we propose DeepMaxent, which harnesses neural networks to automatically learn shared features among species, using the maximum entropy principle. To do so, it employs a normalised Poisson loss where for each species, presence probabilities across sites are modelled by a neural network. We evaluate DeepMaxent on a benchmark dataset known for its spatial sampling biases, using PO data for calibration and presence-absence (PA) data for validation across six regions with different biological groups and covariates. Our results indicate that DeepMaxent performs better than Maxent and other leading SDMs across all regions and taxonomic groups. The method performs particularly well in regions of uneven sampling, demonstrating substantial potential to increase SDM performances. In particular, our approach yields more accurate predictions than traditional single-species models, which opens up new possibilities for methodological enhancement.

Tanguy Lefort, Antoine Affouard, Benjamin Charlier et al.

Deep learning models for plant species identification rely on large annotated datasets. The PlantNet system enables global data collection by allowing users to upload and annotate plant observations, leading to noisy labels due to diverse user skills. Achieving consensus is crucial for training, but the vast scale of collected data makes traditional label aggregation strategies challenging. Existing methods either retain all observations, resulting in noisy training data or selectively keep those with sufficient votes, discarding valuable information. Additionally, as many species are rarely observed, user expertise can not be evaluated as an inter-user agreement: otherwise, botanical experts would have a lower weight in the AI training step than the average user. Our proposed label aggregation strategy aims to cooperatively train plant identification AI models. This strategy estimates user expertise as a trust score per user based on their ability to identify plant species from crowdsourced data. The trust score is recursively estimated from correctly identified species given the current estimated labels. This interpretable score exploits botanical experts' knowledge and the heterogeneity of users. Subsequently, our strategy removes unreliable observations but retains those with limited trusted annotations, unlike other approaches. We evaluate PlantNet's strategy on a released large subset of the PlantNet database focused on European flora, comprising over 6M observations and 800K users. We demonstrate that estimating users' skills based on the diversity of their expertise enhances labeling performance. Our findings emphasize the synergy of human annotation and data filtering in improving AI performance for a refined dataset. We explore incorporating AI-based votes alongside human input. This can further enhance human-AI interactions to detect unreliable observations.

Camille Garcin, Maximilien Servajean, Alexis Joly et al.

In modern classification tasks, the number of labels is getting larger and larger, as is the size of the datasets encountered in practice. As the number of classes increases, class ambiguity and class imbalance become more and more problematic to achieve high top-1 accuracy. Meanwhile, Top-K metrics (metrics allowing K guesses) have become popular, especially for performance reporting. Yet, proposing top-K losses tailored for deep learning remains a challenge, both theoretically and practically. In this paper we introduce a stochastic top-K hinge loss inspired by recent developments on top-K calibrated losses. Our proposal is based on the smoothing of the top-K operator building on the flexible "perturbed optimizer" framework. We show that our loss function performs very well in the case of balanced datasets, while benefiting from a significantly lower computational time than the state-of-the-art top-K loss function. In addition, we propose a simple variant of our loss for the imbalanced case. Experiments on a heavy-tailed dataset show that our loss function significantly outperforms other baseline loss functions.

Titouan Lorieul, Alexis Joly, Dennis Shasha

When many labels are possible, choosing a single one can lead to low precision. A common alternative, referred to as top-$K$ classification, is to choose some number $K$ (commonly around 5) and to return the $K$ labels with the highest scores. Unfortunately, for unambiguous cases, $K>1$ is too many and, for very ambiguous cases, $K \leq 5$ (for example) can be too small. An alternative sensible strategy is to use an adaptive approach in which the number of labels returned varies as a function of the computed ambiguity, but must average to some particular $K$ over all the samples. We denote this alternative average-$K$ classification. This paper formally characterizes the ambiguity profile when average-$K$ classification can achieve a lower error rate than a fixed top-$K$ classification. Moreover, it provides natural estimation procedures for both the fixed-size and the adaptive classifier and proves their consistency. Finally, it reports experiments on real-world image data sets revealing the benefit of average-$K$ classification over top-$K$ in practice. Overall, when the ambiguity is known precisely, average-$K$ is never worse than top-$K$, and, in our experiments, when it is estimated, this also holds.

Daniel Rosendo, Alexandru Costan, Gabriel Antoniu et al.

In more and more application areas, we are witnessing the emergence of complex workflows that combine computing, analytics and learning. They often require a hybrid execution infrastructure with IoT devices interconnected to cloud/HPC systems (aka Computing Continuum). Such workflows are subject to complex constraints and requirements in terms of performance, resource usage, energy consumption and financial costs. This makes it challenging to optimize their configuration and deployment. We propose a methodology to support the optimization of real-life applications on the Edge-to-Cloud Continuum. We implement it as an extension of E2Clab, a previously proposed framework supporting the complete experimental cycle across the Edge-to-Cloud Continuum. Our approach relies on a rigorous analysis of possible configurations in a controlled testbed environment to understand their behaviour and related performance trade-offs. We illustrate our methodology by optimizing Pl@ntNet, a world-wide plant identification application. Our methodology can be generalized to other applications in the Edge-to-Cloud Continuum.

Rafael S. Pereira, Alexis Joly, Patrick Valduriez et al.

Deep learning models have become increasingly useful in many different industries. On the domain of image classification, convolutional neural networks proved the ability to learn robust features for the closed set problem, as shown in many different datasets, such as MNIST FASHIONMNIST, CIFAR10, CIFAR100, and IMAGENET. These approaches use deep neural networks with dense layers with softmax activation functions in order to learn features that can separate classes in a latent space. However, this traditional approach is not useful for identifying classes unseen on the training set, known as the open set problem. A similar problem occurs in scenarios involving learning on small data. To tackle both problems, few-shot learning has been proposed. In particular, metric learning learns features that obey constraints of a metric distance in the latent space in order to perform classification. However, while this approach proves to be useful for the open set problem, current implementation requires pair-wise training, where both positive and negative examples of similar images are presented during the training phase, which limits the applicability of these approaches in large data or large class scenarios given the combinatorial nature of the possible inputs.In this paper, we present a constraint-based approach applied to the representations in the latent space under the normalized softmax loss, proposed by[18]. We experimentally validate the proposed approach for the classification of unseen classes on different datasets using both metric learning and the normalized softmax loss, on disjoint and joint scenarios. Our results show that not only our proposed strategy can be efficiently trained on larger set of classes, as it does not require pairwise learning, but also present better classification results than the metric learning strategies surpassing its accuracy by a significant margin.

Elijah Cole, Benjamin Deneu, Titouan Lorieul et al.

Understanding the geographic distribution of species is a key concern in conservation. By pairing species occurrences with environmental features, researchers can model the relationship between an environment and the species which may be found there. To facilitate research in this area, we present the GeoLifeCLEF 2020 dataset, which consists of 1.9 million species observations paired with high-resolution remote sensing imagery, land cover data, and altitude, in addition to traditional low-resolution climate and soil variables. We also discuss the GeoLifeCLEF 2020 competition, which aims to use this dataset to advance the state-of-the-art in location-based species recommendation.

Julien Herrmann, Olivier Beaumont, Lionel Eyraud-Dubois et al.

This paper introduces a new activation checkpointing method which allows to significantly decrease memory usage when training Deep Neural Networks with the back-propagation algorithm. Similarly to checkpoint-ing techniques coming from the literature on Automatic Differentiation, it consists in dynamically selecting the forward activations that are saved during the training phase, and then automatically recomputing missing activations from those previously recorded. We propose an original computation model that combines two types of activation savings: either only storing the layer inputs, or recording the complete history of operations that produced the outputs (this uses more memory, but requires fewer recomputations in the backward phase), and we provide an algorithm to compute the optimal computation sequence for this model. This paper also describes a PyTorch implementation that processes the entire chain, dealing with any sequential DNN whose internal layers may be arbitrarily complex and automatically executing it according to the optimal checkpointing strategy computed given a memory limit. Through extensive experiments, we show that our implementation consistently outperforms existing checkpoint-ing approaches for a large class of networks, image sizes and batch sizes.

Benjamin Deneu, Maximilien Servajean, Christophe Botella et al.

This paper presents an evaluation of several approaches of plants species distribution modeling based on spatial, environmental and co-occurrences data using machine learning methods. In particular, we re-evaluate the environmental convolutional neural network model that obtained the best performance of the GeoLifeCLEF 2018 challenge but on a revised dataset that fixes some of the issues of the previous one. We also go deeper in the analysis of co-occurrences information by evaluating a new model that jointly takes environmental variables and co-occurrences as inputs of an end-to-end network. Results show that the environmental models are the best performing methods and that there is a significant amount of complementary information between co-occurrences and environment. Indeed, the model learned on both inputs allows a significant performance gain compared to the environmental model alone.