{"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/pvtv2-improved-baselines-with-pyramid-vision","title":"PVT v2: Improved Baselines with Pyramid Vision Transformer","arxiv_id":"2106.13797","date":"2021-06-25","proceeding":null,"authors":["Wenhai Wang","Enze Xie","Xiang Li","Deng-Ping Fan","Kaitao Song","Ding Liang","Tong Lu","Ping Luo","Ling Shao"],"abstract":"Transformer recently has presented encouraging progress in computer vision. In this work, we present new baselines by improving the original Pyramid Vision Transformer (PVT v1) by adding three designs, including (1) linear complexity attention layer, (2) overlapping patch embedding, and (3) convolutional feed-forward network. With these modifications, PVT v2 reduces the computational complexity of PVT v1 to linear and achieves significant improvements on fundamental vision tasks such as classification, detection, and segmentation. Notably, the proposed PVT v2 achieves comparable or better performances than recent works such as Swin Transformer. We hope this work will facilitate state-of-the-art Transformer researches in computer vision. Code is available at https://github.com/whai362/PVT.","url_abs":"https://arxiv.org/abs/2106.13797v7","url_pdf":"https://arxiv.org/pdf/2106.13797v7.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":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/whai362/PVT","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/Owais-Ansari/Unet3plus","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/martinsbruveris/tensorflow-image-models","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/open-mmlab/mmpose","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/rwightman/pytorch-image-models","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/sithu31296/semantic-segmentation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/xiaohu2015/pvt_detectron2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/BR-IDL/PaddleViT/blob/main/image_classification/PVTv2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":null},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/MS-Mind/MS-Code-02/tree/main/configs/pvt","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/MS-Mind/MS-Code-02/tree/main/configs/pvt_v2","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/Mind23-2/MindCode-110","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":{"status":"ok"}},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/PaddlePaddle/PaddleClas","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/justld/PVTV2_paddle","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"paddle","reach":{"status":"ok"}},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/leondgarse/keras_cv_attention_models/tree/main/keras_cv_attention_models/pvt","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":null},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/mindspore-courses/External-Attention-MindSpore/blob/main/model/backbone/PVT.py","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"mindspore","reach":null},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/open-mmlab/mmdetection","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://github.com/shinya7y/UniverseNet","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"pvtv2-improved-baselines-with-pyramid-vision","repo_url":"https://gitlab.com/birder/birder","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"panoptic-segmentation","task_name":"Panoptic Segmentation"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"depthwise-convolution","method_name":"Depthwise Convolution"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"pvtv2","method_name":"PVTv2"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"stochastic-depth","method_name":"Stochastic Depth"},{"method_slug":"swin-transformer","method_name":"Swin Transformer"},{"method_slug":"transformer","method_name":"Transformer"},{"method_slug":"vision-transformer","method_name":"Vision Transformer"}],"datasets_introduced":[],"methods_introduced":[{"slug":"pvtv2","name":"PVTv2","full_name":"Pyramid Vision Transformer v2"}],"results":[{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PVTv2-B4","rank_in_archive_order":391,"of":1060,"metrics":{"GFLOPs":"11.8","Number of params":"82M","Top 1 Accuracy":"83.8%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PVTv2-B3","rank_in_archive_order":457,"of":1060,"metrics":{"GFLOPs":"6.9","Number of params":"45.2M","Top 1 Accuracy":"83.2%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PVTv2-B2","rank_in_archive_order":585,"of":1060,"metrics":{"GFLOPs":"4","Number of params":"25.4M","Top 1 Accuracy":"82%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PVTv2-B1","rank_in_archive_order":813,"of":1060,"metrics":{"GFLOPs":"2.1","Number of params":"13.1M","Top 1 Accuracy":"78.7%"},"uses_additional_data":false},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PVTv2-B0","rank_in_archive_order":1021,"of":1060,"metrics":{"GFLOPs":"0.6","Number of params":"3.4M","Top 1 Accuracy":"70.5%"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-minival","task":"Object Detection","dataset":"COCO minival","model":"Sparse R-CNN (PVTv2-B2)","rank_in_archive_order":80,"of":220,"metrics":{"AP50":"69.5","AP75":"54.9","box AP":"50.1"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-coco-o","task":"Object Detection","dataset":"COCO-O","model":"PVTv2-B5\n(Mask R-CNN)","rank_in_archive_order":23,"of":45,"metrics":{"Average mAP":"28.2","Effective Robustness":"6.85"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2106.13797","atlas_url":"https://app.syntology.ai/?focus=2106.13797","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}