{"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/metasaug-meta-semantic-augmentation-for-long","title":"MetaSAug: Meta Semantic Augmentation for Long-Tailed Visual Recognition","arxiv_id":"2103.12579","date":"2021-03-23","proceeding":"CVPR 2021 1","authors":["Shuang Li","Kaixiong Gong","Chi Harold Liu","Yulin Wang","Feng Qiao","Xinjing Cheng"],"abstract":"Real-world training data usually exhibits long-tailed distribution, where several majority classes have a significantly larger number of samples than the remaining minority classes. This imbalance degrades the performance of typical supervised learning algorithms designed for balanced training sets. In this paper, we address this issue by augmenting minority classes with a recently proposed implicit semantic data augmentation (ISDA) algorithm, which produces diversified augmented samples by translating deep features along many semantically meaningful directions. Importantly, given that ISDA estimates the class-conditional statistics to obtain semantic directions, we find it ineffective to do this on minority classes due to the insufficient training data. To this end, we propose a novel approach to learn transformed semantic directions with meta-learning automatically. In specific, the augmentation strategy during training is dynamically optimized, aiming to minimize the loss on a small balanced validation set, which is approximated via a meta update step. Extensive empirical results on CIFAR-LT-10/100, ImageNet-LT, and iNaturalist 2017/2018 validate the effectiveness of our method.","url_abs":"https://arxiv.org/abs/2103.12579v3","url_pdf":"https://arxiv.org/pdf/2103.12579v3.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":"metasaug-meta-semantic-augmentation-for-long","repo_url":"https://github.com/BIT-DA/MetaSAug","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"long-tail-learning","task_name":"Long-tail Learning"},{"task_slug":"meta-learning","task_name":"Meta-Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-inaturalist","task":"Image Classification","dataset":"iNaturalist","model":"MetaSAug","rank_in_archive_order":16,"of":19,"metrics":{"Top 1 Accuracy":"63.28%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-inaturalist-2018","task":"Image Classification","dataset":"iNaturalist 2018","model":"MetaSAug","rank_in_archive_order":42,"of":60,"metrics":{"Top-1 Accuracy":"68.75%"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-10-lt-r-10","task":"Long-tail Learning","dataset":"CIFAR-10-LT (ρ=10)","model":"MetaSAug-LDAM","rank_in_archive_order":25,"of":50,"metrics":{"Error Rate":"10.32"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-10-lt-r-100","task":"Long-tail Learning","dataset":"CIFAR-10-LT (ρ=100)","model":"MetaSAug-LDAM","rank_in_archive_order":20,"of":28,"metrics":{"Error Rate":"19.34"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-10-lt-r-200","task":"Long-tail Learning","dataset":"CIFAR-10-LT (ρ=200)","model":"MetaSAug-LDAM","rank_in_archive_order":2,"of":2,"metrics":{"Error Rate":"22.65"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-10-lt-r-50","task":"Long-tail Learning","dataset":"CIFAR-10-LT (ρ=50)","model":"MetaSAug-LDAM","rank_in_archive_order":8,"of":8,"metrics":{"Error Rate":"15.66"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-100-lt-r-10","task":"Long-tail Learning","dataset":"CIFAR-100-LT (ρ=10)","model":"MetaSAug-LDAM","rank_in_archive_order":23,"of":31,"metrics":{"Error Rate":"38.72"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-100-lt-r-100","task":"Long-tail Learning","dataset":"CIFAR-100-LT (ρ=100)","model":"MetaSAug-LDAM","rank_in_archive_order":35,"of":66,"metrics":{"Error Rate":"51.99"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-100-lt-r-200","task":"Long-tail Learning","dataset":"CIFAR-100-LT (ρ=200)","model":"MetaSAug-LDAM","rank_in_archive_order":2,"of":2,"metrics":{"Error Rate":"56.91"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-cifar-100-lt-r-50","task":"Long-tail Learning","dataset":"CIFAR-100-LT (ρ=50)","model":"MetaSAug-LDAM","rank_in_archive_order":21,"of":25,"metrics":{"Error Rate":"47.73"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-imagenet-lt","task":"Long-tail Learning","dataset":"ImageNet-LT","model":"MetaSAug (ResNet-152)","rank_in_archive_order":55,"of":69,"metrics":{"Top-1 Accuracy":"50.03"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-imagenet-lt","task":"Long-tail Learning","dataset":"ImageNet-LT","model":"MetaSAug with CE loss","rank_in_archive_order":56,"of":69,"metrics":{"Top-1 Accuracy":"47.39"},"uses_additional_data":false},{"leaderboard":"/sota/long-tail-learning-on-inaturalist-2018","task":"Long-tail Learning","dataset":"iNaturalist 2018","model":"MetaSAug","rank_in_archive_order":38,"of":43,"metrics":{"Top-1 Accuracy":"68.75%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2103.12579","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2103.12579"}},"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":"deterministic:regex_extraction","url":"https://github.com/BIT-DA/MetaSAug","reach":null}],"summary":{"ran":2,"unverified":1},"by_repo_kind":{"official":{"samples":3,"ran":2,"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":"885f4cef722f0ecb","entry":"MetaBatchNorm2d","repo":"BIT-DA/MetaSAug","repo_kind":"official","path":"ImageNet_iNat/resnet_meta.py","file_url":"https://github.com/BIT-DA/MetaSAug/blob/HEAD/ImageNet_iNat/resnet_meta.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"885f4cef722f0ecb"}},{"code_sha256_prefix":"c5563b3215911fe1","entry":"MetaConv2d","repo":"BIT-DA/MetaSAug","repo_kind":"official","path":"ImageNet_iNat/resnet_meta.py","file_url":"https://github.com/BIT-DA/MetaSAug/blob/HEAD/ImageNet_iNat/resnet_meta.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c5563b3215911fe1"}},{"code_sha256_prefix":"9656bad1da39d4b0","entry":"FeatureMeta","repo":"BIT-DA/MetaSAug","repo_kind":"official","path":"ImageNet_iNat/resnet_meta.py","file_url":"https://github.com/BIT-DA/MetaSAug/blob/HEAD/ImageNet_iNat/resnet_meta.py","link_basis":"first_harvest_node","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":"9656bad1da39d4b0"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}