{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,22]],"date-time":"2026-07-22T17:49:06Z","timestamp":1784742546535,"version":"3.55.0"},"reference-count":36,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2024,10,14]],"date-time":"2024-10-14T00:00:00Z","timestamp":1728864000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Neurorobot."],"abstract":"<jats:p>Object detection plays a crucial role in robotic vision, focusing on accurately identifying and localizing objects within images. However, many existing methods encounter limitations, particularly when it comes to effectively implementing a one-to-many matching strategy. To address these challenges, we propose NAN-DETR (Noising Multi-Anchor Detection Transformer), an innovative framework based on DETR (Detection Transformer). NAN-DETR introduces three key improvements to transformer-based object detection: a decoder-based multi-anchor strategy, a centralization noising mechanism, and the integration of Complete Intersection over Union (CIoU) loss. The multi-anchor strategy leverages multiple anchors per object, significantly enhancing detection accuracy by improving the one-to-many matching process. The centralization noising mechanism mitigates conflicts among anchors by injecting controlled noise into the detection boxes, thereby increasing the robustness of the model. Additionally, CIoU loss, which incorporates both aspect ratio and spatial distance in its calculations, results in more precise bounding box predictions compared to the conventional IoU loss. Although NAN-DETR may not drastically improve real-time processing capabilities, its exceptional performance positions it as a highly reliable solution for diverse object detection scenarios.<\/jats:p>","DOI":"10.3389\/fnbot.2024.1484088","type":"journal-article","created":{"date-parts":[[2024,10,14]],"date-time":"2024-10-14T05:10:36Z","timestamp":1728882636000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":7,"title":["NAN-DETR: noising multi-anchor makes DETR better for object detection"],"prefix":"10.3389","volume":"18","author":[{"given":"Zixin","family":"Huang","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xuesong","family":"Tao","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xinyuan","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2024,10,14]]},"reference":[{"key":"B1","first-page":"5562","article-title":"\u201cSoft-nms - improving object detection with one line of code,\u201d","author":"Bodla","year":"2017"},{"key":"B2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_13","article-title":"End-to-end object detection with transformers","author":"Carion","year":"2020","journal-title":"arXiv"},{"key":"B3","doi-asserted-by":"crossref","DOI":"10.1109\/ICCV51070.2023.00610","article-title":"\u201cGroup detr: Fast detr training with group-wise one-to-many assignment,\u201d","volume-title":"Proceedings of the IEEE International Conference on Computer Vision (ICCV)","author":"Chen","year":"2023"},{"key":"B4","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2207.08914","article-title":"Conditional detr v2: Efficient detection transformer with box queries","author":"Chen","year":"2022","journal-title":"arXiv"},{"key":"B5","first-page":"5250","article-title":"\u201cRecurrent glimpse-based decoder for detection with transformer,\u201d","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022","author":"Chen","year":"2022"},{"key":"B6","first-page":"2968","article-title":"\u201cDynamic DETR: end-to-end object detection with dynamic attention,\u201d","volume-title":"2021 IEEE\/CVF International Conference on Computer Vision, ICCV 2021","author":"Dai","year":"2021"},{"key":"B7","doi-asserted-by":"publisher","first-page":"12772","DOI":"10.1109\/TPAMI.2022.3216514","article-title":"Unsupervised pre-training for detection transformers","volume":"45","author":"Dai","year":"2022","journal-title":"IEEE Trans. Pattern Analy. Mach. Intellig"},{"key":"B8","doi-asserted-by":"crossref","DOI":"10.1109\/ICCV48922.2021.00360","article-title":"\u201cFast convergence of DETR with spatially modulated co-attention,\u201d","volume-title":"2021 IEEE\/CVF International Conference on Computer Vision, ICCV 2021","author":"Gao","year":"2021"},{"key":"B9","first-page":"5354","article-title":"\u201cAdamixer: A fast-converging query-based object detector,\u201d","author":"Gao","year":"2022"},{"key":"B10","doi-asserted-by":"crossref","first-page":"580","DOI":"10.1109\/CVPR.2014.81","article-title":"\u201cRich feature hierarchies for accurate object detection and semantic segmentation,\u201d","volume-title":"2014 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2014","author":"Girshick","year":"2014"},{"key":"B11","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR.2016.90","article-title":"\u201cDeep residual learning for image recognition,\u201d","volume-title":"2016 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016","author":"He","year":"2016"},{"key":"B12","doi-asserted-by":"crossref","first-page":"17574","DOI":"10.1109\/CVPR52733.2024.01664","article-title":"\u201cSalience detr: Enhancing detection transformer with hierarchical salience filtering refinement,\u201d","volume-title":"2024 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Hou","year":"2024"},{"key":"B13","first-page":"19702","article-title":"\u201cDetrs with hybrid matching,\u201d","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","author":"Jia","year":"2023"},{"key":"B14","doi-asserted-by":"publisher","first-page":"7389","DOI":"10.1109\/TIP.2020.3002345","article-title":"Foveabox: beyound anchor-based object detection","volume":"29","author":"Kong","year":"2020","journal-title":"IEEE Trans. Image Proc"},{"key":"B15","doi-asserted-by":"publisher","first-page":"83","DOI":"10.1002\/nav.3800020109","article-title":"The hungarian method for the assignment problem","volume":"2","author":"Kuhn","year":"1955","journal-title":"Naval Res. Logist. Quart"},{"key":"B16","first-page":"13609","article-title":"\u201cDN-DETR: accelerate DETR training by introducing query denoising,\u201d","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022","author":"Li","year":"2022"},{"key":"B17","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1405.0312","article-title":"Microsoft coco: Common objects in context","author":"Lin","year":"2015","journal-title":"arXiv"},{"key":"B18","first-page":"2959","article-title":"\u201cWB-DETR: transformer-based detector without backbone,\u201d","volume-title":"2021 IEEE\/CVF International Conference on Computer Vision, ICCV 2021","author":"Liu","year":"2021"},{"key":"B19","article-title":"\u201cDAB-DETR: dynamic anchor boxes are better queries for DETR,\u201d","author":"Liu","year":"2022","journal-title":"The Tenth International Conference on Learning Representations, ICLR 2022"},{"key":"B20","doi-asserted-by":"crossref","DOI":"10.1109\/CVPR52688.2022.01170","article-title":"\u201cSwin transformer V2: scaling up capacity and resolution,\u201d","volume-title":"IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022","author":"Liu","year":"2022"},{"key":"B21","first-page":"9992","article-title":"\u201cSwin transformer: Hierarchical vision transformer using shifted windows,\u201d","volume-title":"2021 IEEE\/CVF International Conference on Computer Vision, ICCV 2021","author":"Liu","year":"2021"},{"key":"B22","first-page":"3631","article-title":"\u201cConditional DETR for fast training convergence,\u201d","volume-title":"2021 IEEE\/CVF International Conference on Computer Vision, ICCV 2021","author":"Meng","year":"2021"},{"key":"B23","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2212.06137","article-title":"NMS strikes back","author":"Ouyang-Zhang","year":"2022","journal-title":"arXiv"},{"key":"B24","first-page":"779","article-title":"\u201cYou only look once: Unified, real-time object detection,\u201d","volume-title":"2016 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016","author":"Redmon","year":"2016"},{"key":"B25","first-page":"91","article-title":"\u201cFaster R-CNN: towards real-time object detection with region proposal networks,\u201d","author":"Ren","year":"2015","journal-title":"Advances in Neural Information Processing Systems"},{"key":"B26","first-page":"658","article-title":"\u201cGeneralized intersection over union: A metric and a loss for bounding box regression,\u201d","volume-title":"IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2019","author":"Rezatofighi","year":"2019"},{"key":"B27","article-title":"\u201cSparse DETR: efficient end-to-end object detection with learnable sparsity,\u201d","volume-title":"The Tenth International Conference on Learning Representations, ICLR 2022","author":"Roh","year":"2022"},{"key":"B28","first-page":"5998","article-title":"\u201cAttention is all you need,\u201d","author":"Vaswani","year":"2017","journal-title":"Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017"},{"key":"B29","doi-asserted-by":"publisher","first-page":"2567","DOI":"10.1609\/aaai.v36i3.20158","article-title":"Anchor detr: query design for transformer-based detector","volume":"36","author":"Wang","year":"2022","journal-title":"Proc. AAAI Conf. Artif. Intellig"},{"key":"B30","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2104.01318","article-title":"Efficient DETR: improving end-to-end object detector with dense prior","author":"Yao","year":"2021","journal-title":"arXiv"},{"key":"B31","article-title":"\u201cDINO: DETR with improved denoising anchor boxes for end-to-end object detection,\u201d","volume-title":"The Eleventh International Conference on Learning Representations","author":"Zhang","year":"2023"},{"key":"B32","first-page":"9756","article-title":"\u201cBridging the gap between anchor-based and anchor-free detection via adaptive training sample selection,\u201d","volume-title":"2020 IEEE\/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2020","author":"Zhang","year":"2020"},{"key":"B33","first-page":"16965","article-title":"\u201cDetrs beat yolos on real-time object detection,\u201d","author":"Zhao","year":"2024","journal-title":"Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"},{"key":"B34","first-page":"12993","article-title":"\u201cDistance-iou loss: Faster and better learning for bounding box regression,\u201d","volume-title":"The Thirty-Fourth AAAI Conference on Artificial Intelligence, AAAI 2020, The Thirty-Second Innovative Applications of Artificial Intelligence Conference, IAAI 2020, The Tenth AAAI Symposium on Educational Advances in Artificial Intelligence, EAAI 2020","author":"Zheng","year":"2020"},{"key":"B35","article-title":"\u201cDeformable DETR: deformable transformers for end-to-end object detection,\u201d","volume-title":"9th International Conference on Learning Representations, ICLR 2021","author":"Zhu","year":"2021"},{"key":"B36","first-page":"6748","article-title":"\u201cDetrs with collaborative hybrid assignments training,\u201d","author":"Zong","year":"2023","journal-title":"Proceedings of the IEEE\/CVF International Conference on Computer Vision"}],"container-title":["Frontiers in Neurorobotics"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2024.1484088\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,10,14]],"date-time":"2024-10-14T05:10:46Z","timestamp":1728882646000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fnbot.2024.1484088\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,10,14]]},"references-count":36,"alternative-id":["10.3389\/fnbot.2024.1484088"],"URL":"https:\/\/doi.org\/10.3389\/fnbot.2024.1484088","relation":{},"ISSN":["1662-5218"],"issn-type":[{"value":"1662-5218","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,10,14]]},"article-number":"1484088"}}