{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/efficient-deep-learning-models-for-land-cover","title":"Benchmarking and scaling of deep learning models for land cover image classification","arxiv_id":"2111.09451","date":"2021-11-18","proceeding":null,"authors":["Ioannis Papoutsis","Nikolaos-Ioannis Bountos","Angelos Zavras","Dimitrios Michail","Christos Tryfonopoulos"],"abstract":"The availability of the sheer volume of Copernicus Sentinel-2 imagery has created new opportunities for exploiting deep learning (DL) methods for land use land cover (LULC) image classification. However, an extensive set of benchmark experiments is currently lacking, i.e. DL models tested on the same dataset, with a common and consistent set of metrics, and in the same hardware. In this work, we use the BigEarthNet Sentinel-2 dataset to benchmark for the first time different state-of-the-art DL models for the multi-label, multi-class LULC image classification problem, contributing with an exhaustive zoo of 60 trained models. Our benchmark includes standard CNNs, as well as non-convolutional methods. We put to the test EfficientNets and Wide Residual Networks (WRN) architectures, and leverage classification accuracy, training time and inference rate. Furthermore, we propose to use the EfficientNet framework for the compound scaling of a lightweight WRN. Enhanced with an Efficient Channel Attention mechanism, our scaled lightweight model emerged as the new state-of-the-art. It achieves 4.5% higher averaged F-Score classification accuracy for all 19 LULC classes compared to a standard ResNet50 baseline model, with an order of magnitude less trainable parameters. We provide access to all trained models, along with our code for distributed training on multiple GPU nodes. This model zoo of pre-trained encoders can be used for transfer learning and rapid prototyping in different remote sensing tasks that use Sentinel-2 data, instead of exploiting backbone models trained with data from a different domain, e.g., from ImageNet. We validate their suitability for transfer learning in different datasets of diverse volumes. Our top-performing WRN achieves state-of-the-art performance (71.1% F-Score) on the SEN12MS dataset while being exposed to only a small fraction of the training dataset.","url_abs":"https://arxiv.org/abs/2111.09451v3","url_pdf":"https://arxiv.org/pdf/2111.09451v3.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"efficient-deep-learning-models-for-land-cover","repo_url":"https://github.com/orion-ai-lab/efficientbigearthnet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"benchmarking","task_name":"Benchmarking"},{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"multi-label-image-classification","task_name":"Multi-Label Image Classification"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"depthwise-convolution","method_name":"Depthwise Convolution"},{"method_slug":"depthwise-separable-convolution","method_name":"Depthwise Separable Convolution"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"ecanet","method_name":"Efficient Channel Attention"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"inverted-residual-block","method_name":"Inverted Residual Block"},{"method_slug":"pointwise-convolution","method_name":"Pointwise Convolution"},{"method_slug":"rmsprop","method_name":"RMSProp"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"squeeze-and-excitation-block","method_name":"Squeeze-and-Excitation Block"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/multi-label-image-classification-on","task":"Multi-Label Image Classification","dataset":"BigEarthNet","model":"WideResNet-B5-ECA","rank_in_archive_order":7,"of":10,"metrics":{"FScore":"79","official split":"Yes"},"uses_additional_data":false},{"leaderboard":"/sota/multi-label-image-classification-on","task":"Multi-Label Image Classification","dataset":"BigEarthNet","model":"ViTM/20","rank_in_archive_order":8,"of":10,"metrics":{"FScore":"77.1","official split":"Yes"},"uses_additional_data":false},{"leaderboard":"/sota/multi-label-image-classification-on","task":"Multi-Label Image Classification","dataset":"BigEarthNet","model":"ResNet50","rank_in_archive_order":9,"of":10,"metrics":{"FScore":"76.8","official split":"Yes"},"uses_additional_data":false},{"leaderboard":"/sota/multi-label-image-classification-on","task":"Multi-Label Image Classification","dataset":"BigEarthNet","model":"MLPMixer","rank_in_archive_order":10,"of":10,"metrics":{"FScore":"75.2","official split":"Yes"},"uses_additional_data":false},{"leaderboard":"/sota/multi-label-image-classification-on-2","task":"Multi-Label Image Classification","dataset":"BigEarthNet (official test set)","model":"ResNet50","rank_in_archive_order":6,"of":6,"metrics":{"F1 Score":"76.8"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2111.09451","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2111.09451"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/orion-ai-lab/efficientbigearthnet","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":5},"by_repo_kind":{"official":{"samples":5,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"a90db59f5000c624","entry":"MlpMixerModel","repo":"orion-ai-lab/efficientbigearthnet","repo_kind":"official","path":"models.py","file_url":"https://github.com/orion-ai-lab/efficientbigearthnet/blob/HEAD/models.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a90db59f5000c624"}},{"code_sha256_prefix":"a54c0fcde85f305b","entry":"coord_module","repo":"orion-ai-lab/efficientbigearthnet","repo_kind":"official","path":"modules.py","file_url":"https://github.com/orion-ai-lab/efficientbigearthnet/blob/HEAD/modules.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a54c0fcde85f305b"}},{"code_sha256_prefix":"305e2195acf9efc5","entry":"create_batched_dataset","repo":"orion-ai-lab/efficientbigearthnet","repo_kind":"official","path":"inputs.py","file_url":"https://github.com/orion-ai-lab/efficientbigearthnet/blob/HEAD/inputs.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"305e2195acf9efc5"}},{"code_sha256_prefix":"6bad926586ca04c0","entry":"eca_module","repo":"orion-ai-lab/efficientbigearthnet","repo_kind":"official","path":"modules.py","file_url":"https://github.com/orion-ai-lab/efficientbigearthnet/blob/HEAD/modules.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6bad926586ca04c0"}},{"code_sha256_prefix":"90e46f4b8d8dd1b9","entry":"ghost_module","repo":"orion-ai-lab/efficientbigearthnet","repo_kind":"official","path":"modules.py","file_url":"https://github.com/orion-ai-lab/efficientbigearthnet/blob/HEAD/modules.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"90e46f4b8d8dd1b9"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}