{"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/pretraining-boosts-out-of-domain-robustness","title":"Pretraining boosts out-of-domain robustness for pose estimation","arxiv_id":"1909.11229","date":"2019-09-24","proceeding":null,"authors":["Alexander Mathis","Thomas Biasi","Steffen Schneider","Mert Yüksekgönül","Byron Rogers","Matthias Bethge","Mackenzie W. Mathis"],"abstract":"Neural networks are highly effective tools for pose estimation. However, as in other computer vision tasks, robustness to out-of-domain data remains a challenge, especially for small training sets that are common for real-world applications. Here, we probe the generalization ability with three architecture classes (MobileNetV2s, ResNets, and EfficientNets) for pose estimation. We developed a dataset of 30 horses that allowed for both \"within-domain\" and \"out-of-domain\" (unseen horse) benchmarking - this is a crucial test for robustness that current human pose estimation benchmarks do not directly address. We show that better ImageNet-performing architectures perform better on both within- and out-of-domain data if they are first pretrained on ImageNet. We additionally show that better ImageNet models generalize better across animal species. Furthermore, we introduce Horse-C, a new benchmark for common corruptions for pose estimation, and confirm that pretraining increases performance in this domain shift context as well. Overall, our results demonstrate that transfer learning is beneficial for out-of-domain robustness.","url_abs":"https://arxiv.org/abs/1909.11229v2","url_pdf":"https://arxiv.org/pdf/1909.11229v2.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":"pretraining-boosts-out-of-domain-robustness","repo_url":"https://github.com/DeepLabCut/DeepLabCut","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"LGPL-3.0"}}],"tasks":[{"task_slug":"animal-pose-estimation","task_name":"Animal Pose Estimation"},{"task_slug":"benchmarking","task_name":"Benchmarking"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"},{"task_slug":"transfer-learning","task_name":"Transfer Learning"}],"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":"efficientnet","method_name":"EfficientNet"},{"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":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"squeeze-and-excitation-block","method_name":"Squeeze-and-Excitation Block"}],"datasets_introduced":[{"slug":"horse-10","name":"Horse-10","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/animal-pose-estimation-on-horse-10","task":"Animal Pose Estimation","dataset":"Horse-10","model":"DeepLabCut-EfficientNet-B6","rank_in_archive_order":1,"of":8,"metrics":{"PCK@0.3 (OOD)":"88.4"},"uses_additional_data":false},{"leaderboard":"/sota/animal-pose-estimation-on-horse-10","task":"Animal Pose Estimation","dataset":"Horse-10","model":"DeepLabCut-EfficientNet-B4","rank_in_archive_order":2,"of":8,"metrics":{"PCK@0.3 (OOD)":" 86.9"},"uses_additional_data":false},{"leaderboard":"/sota/animal-pose-estimation-on-horse-10","task":"Animal Pose Estimation","dataset":"Horse-10","model":"DeepLabCut-RESNET-101","rank_in_archive_order":3,"of":8,"metrics":{"PCK@0.3 (OOD)":"84.3"},"uses_additional_data":false},{"leaderboard":"/sota/animal-pose-estimation-on-horse-10","task":"Animal Pose Estimation","dataset":"Horse-10","model":"DeepLabCut-RESNET 50","rank_in_archive_order":4,"of":8,"metrics":{"PCK@0.3 (OOD)":"81.3"},"uses_additional_data":false},{"leaderboard":"/sota/animal-pose-estimation-on-horse-10","task":"Animal Pose Estimation","dataset":"Horse-10","model":"DeepLabCut-MOBILENETV2-1","rank_in_archive_order":5,"of":8,"metrics":{"PCK@0.3 (OOD)":"77.6"},"uses_additional_data":false},{"leaderboard":"/sota/animal-pose-estimation-on-horse-10","task":"Animal Pose Estimation","dataset":"Horse-10","model":"DeepLabCut-MOBILENETV2 0.35","rank_in_archive_order":6,"of":8,"metrics":{"PCK@0.3 (OOD)":"63.5"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/1909.11229","atlas_url":"https://app.syntology.ai/?focus=1909.11229","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}