MonOri: Orientation-Guided PnP for Monocular 3-D Object Detection
Monocular 3-D object detection is a challenging task in the field of autonomous driving and has made great progress. However, current monocular image methods tend to incorporate additional information such as pseudolabels to improve algorithm performance while overlooking the geometric relationship...
Uložené v:
| Vydané v: | IEEE transaction on neural networks and learning systems Ročník 36; číslo 10; s. 19068 - 19080 |
|---|---|
| Hlavní autori: | , , , , , , , , , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
United States
IEEE
01.10.2025
|
| Predmet: | |
| ISSN: | 2162-237X, 2162-2388, 2162-2388 |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | Monocular 3-D object detection is a challenging task in the field of autonomous driving and has made great progress. However, current monocular image methods tend to incorporate additional information such as pseudolabels to improve algorithm performance while overlooking the geometric relationship between the object's keypoints, resulting in low performance for occluded object detection. To address this issue, we find that introducing the orientation information of objects in the 3-D detection pipeline can help improve the detection performance of occluded objects. An orientation-guided perspective-n-point (PnP) for monocular 3-D object detection method named MonOri is presented in this article, which uses object's orientation to guide keypoints' optimization. Considering the existence of different deformation objects in the scene, we design the feature aggregation detection module (FADM), which consists of the feature focus fusion module (FFFM) and CondConv detection module (CCDM). First, FFFM can highlight signals from irregularly occluded objects, effectively modeling features of elongated and small-sized objects. This module enhances the model's ability to recognize elongated and small-sized objects in complex scenes. Then, the CCDM is designed to improve the network's ability to estimate object keypoints' location regression under occlusion conditions and minimize the network computational overhead. Finally, considering that the unoccluded portions of occluded objects are closely related to the orientation of the objects, an orientation-guided keypoints' selection module (OGKSM) is proposed to enhance the accuracy of objected optimization for keypoint positions and spatial location inference of the object. Experimental results indicate that the MonOri method achieves competitive results; it is also demonstrated that the orientation information is introduced in the PnP algorithm to estimate the object's spatial position that can mitigate the impact of occlusion on object detection, thus improving the recognition rate of occluded objects. Our code is available at https://github.com/DL-YHD/MonOri |
|---|---|
| AbstractList | Monocular 3-D object detection is a challenging task in the field of autonomous driving and has made great progress. However, current monocular image methods tend to incorporate additional information such as pseudolabels to improve algorithm performance while overlooking the geometric relationship between the object's keypoints, resulting in low performance for occluded object detection. To address this issue, we find that introducing the orientation information of objects in the 3-D detection pipeline can help improve the detection performance of occluded objects. An orientation-guided perspective-n-point (PnP) for monocular 3-D object detection method named MonOri is presented in this article, which uses object's orientation to guide keypoints' optimization. Considering the existence of different deformation objects in the scene, we design the feature aggregation detection module (FADM), which consists of the feature focus fusion module (FFFM) and CondConv detection module (CCDM). First, FFFM can highlight signals from irregularly occluded objects, effectively modeling features of elongated and small-sized objects. This module enhances the model's ability to recognize elongated and small-sized objects in complex scenes. Then, the CCDM is designed to improve the network's ability to estimate object keypoints' location regression under occlusion conditions and minimize the network computational overhead. Finally, considering that the unoccluded portions of occluded objects are closely related to the orientation of the objects, an orientation-guided keypoints' selection module (OGKSM) is proposed to enhance the accuracy of objected optimization for keypoint positions and spatial location inference of the object. Experimental results indicate that the MonOri method achieves competitive results; it is also demonstrated that the orientation information is introduced in the PnP algorithm to estimate the object's spatial position that can mitigate the impact of occlusion on object detection, thus improving the recognition rate of occluded objects. Our code is available at https://github.com/DL-YHD/MonOri Monocular 3-D object detection is a challenging task in the field of autonomous driving and has made great progress. However, current monocular image methods tend to incorporate additional information such as pseudolabels to improve algorithm performance while overlooking the geometric relationship between the object's keypoints, resulting in low performance for occluded object detection. To address this issue, we find that introducing the orientation information of objects in the 3-D detection pipeline can help improve the detection performance of occluded objects. An orientation-guided perspective-n-point (PnP) for monocular 3-D object detection method named MonOri is presented in this article, which uses object's orientation to guide keypoints' optimization. Considering the existence of different deformation objects in the scene, we design the feature aggregation detection module (FADM), which consists of the feature focus fusion module (FFFM) and CondConv detection module (CCDM). First, FFFM can highlight signals from irregularly occluded objects, effectively modeling features of elongated and small-sized objects. This module enhances the model's ability to recognize elongated and small-sized objects in complex scenes. Then, the CCDM is designed to improve the network's ability to estimate object keypoints' location regression under occlusion conditions and minimize the network computational overhead. Finally, considering that the unoccluded portions of occluded objects are closely related to the orientation of the objects, an orientation-guided keypoints' selection module (OGKSM) is proposed to enhance the accuracy of objected optimization for keypoint positions and spatial location inference of the object. Experimental results indicate that the MonOri method achieves competitive results; it is also demonstrated that the orientation information is introduced in the PnP algorithm to estimate the object's spatial position that can mitigate the impact of occlusion on object detection, thus improving the recognition rate of occluded objects. Our code is available at https://github.com/DL-YHD/MonOri.Monocular 3-D object detection is a challenging task in the field of autonomous driving and has made great progress. However, current monocular image methods tend to incorporate additional information such as pseudolabels to improve algorithm performance while overlooking the geometric relationship between the object's keypoints, resulting in low performance for occluded object detection. To address this issue, we find that introducing the orientation information of objects in the 3-D detection pipeline can help improve the detection performance of occluded objects. An orientation-guided perspective-n-point (PnP) for monocular 3-D object detection method named MonOri is presented in this article, which uses object's orientation to guide keypoints' optimization. Considering the existence of different deformation objects in the scene, we design the feature aggregation detection module (FADM), which consists of the feature focus fusion module (FFFM) and CondConv detection module (CCDM). First, FFFM can highlight signals from irregularly occluded objects, effectively modeling features of elongated and small-sized objects. This module enhances the model's ability to recognize elongated and small-sized objects in complex scenes. Then, the CCDM is designed to improve the network's ability to estimate object keypoints' location regression under occlusion conditions and minimize the network computational overhead. Finally, considering that the unoccluded portions of occluded objects are closely related to the orientation of the objects, an orientation-guided keypoints' selection module (OGKSM) is proposed to enhance the accuracy of objected optimization for keypoint positions and spatial location inference of the object. Experimental results indicate that the MonOri method achieves competitive results; it is also demonstrated that the orientation information is introduced in the PnP algorithm to estimate the object's spatial position that can mitigate the impact of occlusion on object detection, thus improving the recognition rate of occluded objects. Our code is available at https://github.com/DL-YHD/MonOri. Monocular 3-D object detection is a challenging task in the field of autonomous driving and has made great progress. However, current monocular image methods tend to incorporate additional information such as pseudolabels to improve algorithm performance while overlooking the geometric relationship between the object's keypoints, resulting in low performance for occluded object detection. To address this issue, we find that introducing the orientation information of objects in the 3-D detection pipeline can help improve the detection performance of occluded objects. An orientation-guided perspective-n-point (PnP) for monocular 3-D object detection method named MonOri is presented in this article, which uses object's orientation to guide keypoints' optimization. Considering the existence of different deformation objects in the scene, we design the feature aggregation detection module (FADM), which consists of the feature focus fusion module (FFFM) and CondConv detection module (CCDM). First, FFFM can highlight signals from irregularly occluded objects, effectively modeling features of elongated and small-sized objects. This module enhances the model's ability to recognize elongated and small-sized objects in complex scenes. Then, the CCDM is designed to improve the network's ability to estimate object keypoints' location regression under occlusion conditions and minimize the network computational overhead. Finally, considering that the unoccluded portions of occluded objects are closely related to the orientation of the objects, an orientation-guided keypoints' selection module (OGKSM) is proposed to enhance the accuracy of objected optimization for keypoint positions and spatial location inference of the object. Experimental results indicate that the MonOri method achieves competitive results; it is also demonstrated that the orientation information is introduced in the PnP algorithm to estimate the object's spatial position that can mitigate the impact of occlusion on object detection, thus improving the recognition rate of occluded objects. Our code is available at https://github.com/DL-YHD/MonOri. |
| Author | Qiu, Yansheng Cao, Chenglong wang, Yimin Wang, Zheng Chen, Jun Chai, Xiaoyu Yao, Hongdou Han, Pengfei Wang, Xiao Jin, Wei |
| Author_xml | – sequence: 1 givenname: Hongdou orcidid: 0000-0002-6297-7138 surname: Yao fullname: Yao, Hongdou organization: College of Electronic Engineering, National University of Defense Technology, Hefei, China – sequence: 2 givenname: Pengfei orcidid: 0000-0003-3724-3551 surname: Han fullname: Han, Pengfei organization: School of Cybersecurity and the School of Artificial Intelligence, Optics and Electronics (iOPEN), Northwestern Polytechnical University, Xi'an, Shaanxi, China – sequence: 3 givenname: Jun orcidid: 0000-0003-1376-0167 surname: Chen fullname: Chen, Jun email: chenj.whu@gmail.com organization: National Engineering Research Center for Multimedia Software, School of Computer Science, Wuhan University, Wuhan, China – sequence: 4 givenname: Zheng orcidid: 0000-0003-3846-9157 surname: Wang fullname: Wang, Zheng email: wangzwhu@whu.edu.cn organization: National Engineering Research Center for Multimedia Software, School of Computer Science, Wuhan University, Wuhan, China – sequence: 5 givenname: Yansheng orcidid: 0000-0002-5619-0902 surname: Qiu fullname: Qiu, Yansheng organization: National Engineering Research Center for Multimedia Software, School of Computer Science, Wuhan University, Wuhan, China – sequence: 6 givenname: Xiao surname: Wang fullname: Wang, Xiao organization: School of Computer Science and Technology, Wuhan University of Science and Technology, Wuhan, China – sequence: 7 givenname: Yimin surname: wang fullname: wang, Yimin organization: School of Information Engineering, Zhongnan University of Economics and Law, Wuhan, China – sequence: 8 givenname: Xiaoyu orcidid: 0000-0003-2278-1955 surname: Chai fullname: Chai, Xiaoyu organization: Hubei Key Laboratory of Multimedia and Network Communication Engineering, Wuhan University, Wuhan, China – sequence: 9 givenname: Chenglong surname: Cao fullname: Cao, Chenglong organization: College of Electronic Engineering, National University of Defense Technology, Hefei, China – sequence: 10 givenname: Wei surname: Jin fullname: Jin, Wei organization: College of Electronic Engineering, National University of Defense Technology, Hefei, China |
| BackLink | https://www.ncbi.nlm.nih.gov/pubmed/40577294$$D View this record in MEDLINE/PubMed |
| BookMark | eNpFkE9PwkAQxTcGI6h8AWNMj16K-6_d1hsBRRMEEjHxttnuTpOS0sXd9uC3dxHEOcybw-9NZt4l6jW2AYRuCB4RgvOH9WIxfx9RTJMRS4RISXaGBpSkNKYsy3qnWXz20dD7DQ6V4iTl-QXqcxwsNOcDNH6zzdJVj1Fo0LSqrWwTz7rKgIlWzSoqrYsCYnVXKxexeBotiw3oNppCGyTQ1-i8VLWH4VGv0Mfz03ryEs-Xs9fJeB5rKkQbJ8oITFJR4BxKAwQUT41WgmuTsdKQNAFOMC1woXFWUl1QlemMCmbycDTH7ArdH_bunP3qwLdyW3kNda0asJ2XjFKec8EoCejdEe2KLRi5c9VWuW_593YA6AHQznrvoDwhBMt9vPI3XrmPVx7jDabbg6kCgH8DwQnHgrAfHDtz6w |
| CODEN | ITNNAL |
| Cites_doi | 10.1109/CVPR52688.2022.00114 10.1007/978-3-030-58592-1_9 10.1109/CVPRW50498.2020.00506 10.1109/CVPR.2018.00255 10.1109/ICCV48922.2021.01489 10.1007/978-3-031-20074-8_41 10.1109/CVPR.2018.00214 10.1109/CVPR52688.2022.00096 10.1109/CVPR46437.2021.00469 10.1109/LRA.2022.3191849 10.1109/TNNLS.2022.3201534 10.1109/ICCV48922.2021.00313 10.1007/978-3-030-58601-0_19 10.1109/CVPR52733.2024.00976 10.1109/ICRA48891.2023.10160483 10.1609/aaai.v36i2.20074 10.1007/978-3-031-20077-9_23 10.1109/CVPR.2012.6248074 10.1109/WACV57701.2024.00737 10.1109/TITS.2025.3525772 10.1109/CVPR42600.2020.01169 10.1109/ICCV.2019.00695 10.1109/CVPR52733.2024.01601 10.1109/CVPR52733.2024.00979 10.1109/ICCV51070.2023.01815 10.1109/CVPR46437.2021.00845 10.1109/TRO.2022.3204509 10.1109/CVPR46437.2021.00330 10.1007/978-3-031-19769-7_8 10.1177/0278364913491297 10.1109/ICCV51070.2023.00592 10.1109/CVPR52688.2022.00281 10.1109/CVPR52688.2022.00398 10.1109/ICCV.2019.00938 10.1109/ICCVW54120.2021.00107 10.1109/LRA.2023.3238137 10.1109/3DV62453.2024.00088 10.1109/CVPR.2019.00864 10.1109/TIP.2025.3544880 10.1109/TNNLS.2023.3339889 10.1109/TNNLS.2022.3171553 10.1007/978-3-030-58580-8_38 10.1109/TITS.2023.3249909 10.1109/CVPR.2018.00249 10.48550/arXiv.1503.02531 10.1145/3442381.3450078 10.1109/ICCVW.2019.00114 10.1007/978-3-031-19769-7_5 10.1109/CVPR52729.2023.01264 10.1007/978-3-030-58542-6_46 10.1109/TIP.2023.3343112 10.1109/ICCV51070.2023.00840 10.1109/ICCV48922.2021.00321 10.1109/CVPR42600.2020.01211 10.1109/TNNLS.2023.3309809 10.1109/CVPR46437.2021.00052 10.1109/TNNLS.2023.3341097 10.1109/CVPR46437.2021.01024 10.1109/CVPR52688.2022.00173 10.1007/s10846-023-02027-6 10.1109/CVPR46437.2021.00886 10.1609/aaai.v37i3.25391 10.1109/CVPR52729.2023.01678 10.1109/TMI.2023.3237183 10.1109/ICCV51070.2023.00558 10.1109/LRA.2025.3544487 10.1109/ICCV48922.2021.00271 10.1109/LRA.2021.3061343 10.1007/978-3-031-73195-2_22 10.1109/ICCV48922.2021.01535 10.1109/CVPR52729.2023.00121 10.1109/ICEICT.2019.8846270 10.48550/ARXIV.1706.03762 10.1109/CVPR42600.2020.00592 10.1109/ICCV51070.2023.00627 10.1007/978-3-031-72761-0_21 10.1109/ICCV.2019.00208 10.1109/TPAMI.2024.3421277 10.3233/FAIA230555 10.1007/978-3-030-58452-8_13 10.1109/TCSVT.2023.3270728 10.1109/ICCV48922.2021.00310 10.1109/CVPR46437.2021.00747 10.1007/978-3-031-20080-9_6 |
| ContentType | Journal Article |
| DBID | 97E RIA RIE AAYXX CITATION NPM 7X8 |
| DOI | 10.1109/TNNLS.2025.3577618 |
| DatabaseName | IEEE Xplore (IEEE) IEEE All-Society Periodicals Package (ASPP) 1998–Present IEEE Electronic Library (IEL) CrossRef PubMed MEDLINE - Academic |
| DatabaseTitle | CrossRef PubMed MEDLINE - Academic |
| DatabaseTitleList | MEDLINE - Academic PubMed |
| Database_xml | – sequence: 1 dbid: NPM name: PubMed url: http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=PubMed sourceTypes: Index Database – sequence: 2 dbid: RIE name: IEEE Xplore url: https://ieeexplore.ieee.org/ sourceTypes: Publisher – sequence: 3 dbid: 7X8 name: MEDLINE - Academic url: https://search.proquest.com/medline sourceTypes: Aggregation Database |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Computer Science |
| EISSN | 2162-2388 |
| EndPage | 19080 |
| ExternalDocumentID | 40577294 10_1109_TNNLS_2025_3577618 11054071 |
| Genre | orig-research Journal Article |
| GrantInformation_xml | – fundername: Military Equipment Comprehensive Research Project grantid: KY24N036; KY24N018 – fundername: National Natural Science Foundation of China; National Nature Science Foundation of China grantid: 62071388; 62171325; 62302351 funderid: 10.13039/501100001809 – fundername: Youth Autonomous Innovation Science Foundation of National University of Defense Technology grantid: ZK24-43 funderid: 10.13039/501100001809 |
| GroupedDBID | 0R~ 4.4 5VS 6IK 97E AAJGR AASAJ AAWTH ABAZT ABQJQ ABVLG ACIWK ACPRK AENEX AFRAH AGQYO AGSQL AHBIQ AKJIK AKQYR ALMA_UNASSIGNED_HOLDINGS ATWAV BEFXN BFFAM BGNUA BKEBE BPEOZ EBS EJD IFIPE IPLJI JAVBF M43 MS~ O9- OCL PQQKQ RIA RIE RNS AAYXX CITATION NPM 7X8 |
| ID | FETCH-LOGICAL-c277t-5ad70167b09efde1ea46dca74cd83fd165e4102b0bc08f2cb2a8c8273d9564403 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 0 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001518837800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 2162-237X 2162-2388 |
| IngestDate | Thu Oct 02 22:29:40 EDT 2025 Thu Oct 09 01:30:56 EDT 2025 Sat Nov 29 07:16:33 EST 2025 Wed Oct 15 14:20:22 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | true |
| Issue | 10 |
| Language | English |
| License | https://ieeexplore.ieee.org/Xplorehelp/downloads/license-information/IEEE.html https://doi.org/10.15223/policy-029 https://doi.org/10.15223/policy-037 |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-c277t-5ad70167b09efde1ea46dca74cd83fd165e4102b0bc08f2cb2a8c8273d9564403 |
| Notes | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
| ORCID | 0000-0003-3846-9157 0000-0002-5619-0902 0000-0003-1376-0167 0000-0003-3724-3551 0000-0003-2278-1955 0000-0002-6297-7138 |
| PMID | 40577294 |
| PQID | 3224947321 |
| PQPubID | 23479 |
| PageCount | 13 |
| ParticipantIDs | pubmed_primary_40577294 crossref_primary_10_1109_TNNLS_2025_3577618 ieee_primary_11054071 proquest_miscellaneous_3224947321 |
| PublicationCentury | 2000 |
| PublicationDate | 2025-10-01 |
| PublicationDateYYYYMMDD | 2025-10-01 |
| PublicationDate_xml | – month: 10 year: 2025 text: 2025-10-01 day: 01 |
| PublicationDecade | 2020 |
| PublicationPlace | United States |
| PublicationPlace_xml | – name: United States |
| PublicationTitle | IEEE transaction on neural networks and learning systems |
| PublicationTitleAbbrev | TNNLS |
| PublicationTitleAlternate | IEEE Trans Neural Netw Learn Syst |
| PublicationYear | 2025 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| References | ref13 ref57 ref12 ref56 ref15 ref59 ref14 ref58 ref53 Yang (ref62) ref52 ref11 ref55 Cao (ref94) 2024 ref10 ref54 ref17 ref16 ref19 ref18 Cao (ref91) ref93 Lu (ref67) 2023 ref92 ref51 ref50 Wang (ref26) ref90 ref46 ref45 ref89 ref48 ref47 ref42 ref41 ref85 ref44 ref88 ref43 ref87 Zhou (ref39) ref8 ref7 ref9 ref4 ref3 ref6 ref5 ref82 ref81 ref40 ref84 ref83 ref80 ref35 ref79 ref34 ref78 ref37 ref36 ref31 ref75 ref30 ref74 ref33 ref77 ref32 ref76 ref1 ref38 Li (ref86) 2023 ref71 ref70 ref73 ref72 ref24 ref68 ref23 ref25 ref20 ref64 Jinrang (ref69); 36 ref63 ref22 ref66 ref21 ref65 Tan (ref2) 2024; 4 ref28 ref27 ref29 ref60 ref61 Chong (ref49) 2022 |
| References_xml | – ident: ref23 doi: 10.1109/CVPR52688.2022.00114 – ident: ref28 doi: 10.1007/978-3-030-58592-1_9 – ident: ref20 doi: 10.1109/CVPRW50498.2020.00506 – ident: ref59 doi: 10.1109/CVPR.2018.00255 – ident: ref76 doi: 10.1109/ICCV48922.2021.01489 – ident: ref53 doi: 10.1007/978-3-031-20074-8_41 – ident: ref32 doi: 10.1109/CVPR.2018.00214 – year: 2024 ident: ref94 article-title: Collaborative novel object discovery and box-guided cross-modal alignment for open-vocabulary 3D object detection publication-title: arXiv:2406.00830 – ident: ref36 doi: 10.1109/CVPR52688.2022.00096 – ident: ref21 doi: 10.1109/CVPR46437.2021.00469 – ident: ref48 doi: 10.1109/LRA.2022.3191849 – year: 2022 ident: ref49 article-title: MonoDistill: Learning spatial features for monocular 3D object detection publication-title: arXiv:2201.10830 – ident: ref7 doi: 10.1109/TNNLS.2022.3201534 – ident: ref33 doi: 10.1109/ICCV48922.2021.00313 – ident: ref50 doi: 10.1007/978-3-030-58601-0_19 – ident: ref79 doi: 10.1109/CVPR52733.2024.00976 – ident: ref71 doi: 10.1109/ICRA48891.2023.10160483 – ident: ref24 doi: 10.1609/aaai.v36i2.20074 – ident: ref65 doi: 10.1007/978-3-031-20077-9_23 – ident: ref63 doi: 10.1109/CVPR.2012.6248074 – ident: ref51 doi: 10.1109/WACV57701.2024.00737 – ident: ref81 doi: 10.1109/TITS.2025.3525772 – ident: ref9 doi: 10.1109/CVPR42600.2020.01169 – ident: ref18 doi: 10.1109/ICCV.2019.00695 – start-page: 1475 volume-title: Proc. Conf. Robot Learn. ident: ref26 article-title: Probabilistic and geometric depth: Detecting objects in perspective – ident: ref73 doi: 10.1109/CVPR52733.2024.01601 – ident: ref72 doi: 10.1109/CVPR52733.2024.00979 – volume: 4 start-page: 41 issue: 1 year: 2024 ident: ref2 article-title: Integrating advanced computer vision and AI algorithms for autonomous driving systems publication-title: J. Theory Pract. Eng. Sci. – ident: ref4 doi: 10.1109/ICCV51070.2023.01815 – ident: ref10 doi: 10.1109/CVPR46437.2021.00845 – ident: ref11 doi: 10.1109/TRO.2022.3204509 – ident: ref14 doi: 10.1109/CVPR46437.2021.00330 – ident: ref58 doi: 10.1007/978-3-031-19769-7_8 – ident: ref64 doi: 10.1177/0278364913491297 – ident: ref88 doi: 10.1109/ICCV51070.2023.00592 – ident: ref66 doi: 10.1109/CVPR52688.2022.00281 – ident: ref40 doi: 10.1109/CVPR52688.2022.00398 – ident: ref8 doi: 10.1109/ICCV.2019.00938 – ident: ref22 doi: 10.1109/ICCVW54120.2021.00107 – ident: ref70 doi: 10.1109/LRA.2023.3238137 – ident: ref68 doi: 10.1109/3DV62453.2024.00088 – ident: ref19 doi: 10.1109/CVPR.2019.00864 – ident: ref80 doi: 10.1109/TIP.2025.3544880 – ident: ref5 doi: 10.1109/TNNLS.2023.3339889 – ident: ref3 doi: 10.1109/TNNLS.2022.3171553 – ident: ref13 doi: 10.1007/978-3-030-58580-8_38 – ident: ref89 doi: 10.1109/TITS.2023.3249909 – ident: ref35 doi: 10.1109/CVPR.2018.00249 – ident: ref44 doi: 10.48550/arXiv.1503.02531 – ident: ref60 doi: 10.1145/3442381.3450078 – ident: ref29 doi: 10.1109/ICCVW.2019.00114 – volume: 36 start-page: 1 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref69 article-title: Monouni: A unified vehicle and infrastructure-side monocular 3D object detection network with sufficient depth clues – ident: ref77 doi: 10.1007/978-3-031-19769-7_5 – ident: ref85 doi: 10.1109/CVPR52729.2023.01264 – start-page: 2033 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref39 article-title: MoGDE: Boosting mobile monocular 3D object detection with ground depth estimation – ident: ref74 doi: 10.1007/978-3-030-58542-6_46 – ident: ref46 doi: 10.1109/TIP.2023.3343112 – ident: ref41 doi: 10.1109/ICCV51070.2023.00840 – ident: ref31 doi: 10.1109/ICCV48922.2021.00321 – ident: ref25 doi: 10.1109/CVPR42600.2020.01211 – ident: ref1 doi: 10.1109/TNNLS.2023.3309809 – year: 2023 ident: ref67 article-title: GUPNet++: Geometry uncertainty propagation network for monocular 3D object detection publication-title: arXiv:2310.15624 – ident: ref34 doi: 10.1109/CVPR46437.2021.00052 – ident: ref6 doi: 10.1109/TNNLS.2023.3341097 – ident: ref17 doi: 10.1109/CVPR46437.2021.01024 – ident: ref54 doi: 10.1109/CVPR52688.2022.00173 – ident: ref78 doi: 10.1007/s10846-023-02027-6 – ident: ref75 doi: 10.1109/CVPR46437.2021.00886 – ident: ref42 doi: 10.1609/aaai.v37i3.25391 – ident: ref43 doi: 10.1109/CVPR52729.2023.01678 – ident: ref47 doi: 10.1109/TMI.2023.3237183 – ident: ref61 doi: 10.1109/ICCV51070.2023.00558 – ident: ref82 doi: 10.1109/LRA.2025.3544487 – start-page: 219 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref91 article-title: CoDA: Collaborative novel box discovery and cross-modal alignment for open-vocabulary 3D object detection – start-page: 1305 volume-title: Proc. Adv. Neural Inf. Process. Syst. ident: ref62 article-title: CondConv: Conditionally parameterized convolutions for efficient inference – ident: ref83 doi: 10.1109/ICCV48922.2021.00271 – year: 2023 ident: ref86 article-title: Augment and criticize: Exploring informative samples for semi-supervised monocular 3D object detection publication-title: arXiv:2303.11243 – ident: ref52 doi: 10.1109/LRA.2021.3061343 – ident: ref93 doi: 10.1007/978-3-031-73195-2_22 – ident: ref15 doi: 10.1109/ICCV48922.2021.01535 – ident: ref90 doi: 10.1109/CVPR52729.2023.00121 – ident: ref56 doi: 10.1109/ICEICT.2019.8846270 – ident: ref38 doi: 10.48550/ARXIV.1706.03762 – ident: ref30 doi: 10.1109/CVPR42600.2020.00592 – ident: ref84 doi: 10.1109/ICCV51070.2023.00627 – ident: ref92 doi: 10.1007/978-3-031-72761-0_21 – ident: ref16 doi: 10.1109/ICCV.2019.00208 – ident: ref45 doi: 10.1109/TPAMI.2024.3421277 – ident: ref87 doi: 10.3233/FAIA230555 – ident: ref37 doi: 10.1007/978-3-030-58452-8_13 – ident: ref57 doi: 10.1109/TCSVT.2023.3270728 – ident: ref12 doi: 10.1109/ICCV48922.2021.00310 – ident: ref27 doi: 10.1109/CVPR46437.2021.00747 – ident: ref55 doi: 10.1007/978-3-031-20080-9_6 |
| SSID | ssj0000605649 |
| Score | 2.4818885 |
| Snippet | Monocular 3-D object detection is a challenging task in the field of autonomous driving and has made great progress. However, current monocular image methods... |
| SourceID | proquest pubmed crossref ieee |
| SourceType | Aggregation Database Index Database Publisher |
| StartPage | 19068 |
| SubjectTerms | 3-D object detection Accuracy Computational modeling Convolution Data models Depth measurement Feature extraction keypoints’ selection monocular image Multimedia communication Object detection occlusion object orientation guided Prediction algorithms Three-dimensional displays |
| Title | MonOri: Orientation-Guided PnP for Monocular 3-D Object Detection |
| URI | https://ieeexplore.ieee.org/document/11054071 https://www.ncbi.nlm.nih.gov/pubmed/40577294 https://www.proquest.com/docview/3224947321 |
| Volume | 36 |
| WOSCitedRecordID | wos001518837800001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVIEE databaseName: IEEE Xplore customDbUrl: eissn: 2162-2388 dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0000605649 issn: 2162-237X databaseCode: RIE dateStart: 20120101 isFulltext: true titleUrlDefault: https://ieeexplore.ieee.org/ providerName: IEEE |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV07T8MwELZoxcBCeRQoj8pIbChtmjixw1ZRCgNqK1GkbpEfF6lLgvrg93N2kiKGDmwZzk7k75z7zr4HIQ_oxkoXKhWH0hbV9pmnFGSeYUkMnGeGZ5lrNsEnE7FYJLMqWd3lwgCACz6Dnn10d_mm0Ft7VNZHU-XqxTVIg_O4TNbaHaj4SMxjR3eDQRx4QcgXdZKMn_Tnk8n7B7qDQdQLIxw8sI36LFlBcsn-2CTXZGU_33R2Z9z65xefkOOKYNJhqRGn5ADyM9KqmzfQai-fkyHu5ulq-USnq2WVf5R7r9ulAUNn-YwimaUoUrg4VRp6IzpV9syGjmDjwrfyNvkcv8yf37yqn4KnA843XiQNt1kHyk8gMzAAyWKjJWfaiDAzgzgChnxD-Ur7Igu0CqTQAvmNQSeKMT-8IM28yOGKUB0htkpmUiSCgY6lrTsmEqmRDioDqkMe6xVNv8qyGalzN_wkdVCkFoq0gqJD2nbpfiWrVeuQ-xqFFLXeXmXIHIrtOsXfEEsYDwOUuSzh2Y2uUb3eM-sNObIvLyPybklzs9rCHTnU35vletVF1VqIrlOtH8srxwo |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LT9wwEB4VigQXXqVloVBX4oYCXseJE24IClRssyuxlfYW-TGR9pJFyy6_n7GTgDhw4JbDJLLmG2e-secBcEJhrA6pUmmsfVNtLiNjsIqczFNUqnKqqsKwCVUU2WSSj9pi9VALg4gh-QzP_GO4y3czu_RHZefkqkK_uBX4mkgpeFOu9Xqkwomap4Hwin4qIhGrSVcmw_PzcVEMHiggFMlZnCiK3v2oPk9XiF7Kd14pjFn5mHEGz3Oz9ck1b8NmSzHZZWMTO_AF613Y6sY3sHY3f4NL2s_D-fSCDefTtgKpjm6XU4eOjeoRIzrLSGQWMlVZHF2zofGnNuwaFyGBq96D_zd_xld3UTtRIbJCqUWUaKd83YHhOVYO-6hl6qxW0rosrlw_TVAS4zDcWJ5VwhqhM5sRw3EURknJ4--wWs9q3AdmE0LX6EpneSbRptp3HstybYkQGoemB6edRsvHpnFGGQIOnpcBitJDUbZQ9GDPq-5NstVaD353KJRk9_4yQ9c4Wz6V9COSuVSxIJkfDTyvb3eoHnzw1V-wfjf-NygHf4v7Q9jwC2ny837C6mK-xCNYs8-L6dP8OBjYC6I3yWk |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=MonOri%3A+Orientation-Guided+PnP+for+Monocular+3-D+Object+Detection&rft.jtitle=IEEE+transaction+on+neural+networks+and+learning+systems&rft.au=Yao%2C+Hongdou&rft.au=Han%2C+Pengfei&rft.au=Chen%2C+Jun&rft.au=Wang%2C+Zheng&rft.date=2025-10-01&rft.pub=IEEE&rft.issn=2162-237X&rft.volume=36&rft.issue=10&rft.spage=19068&rft.epage=19080&rft_id=info:doi/10.1109%2FTNNLS.2025.3577618&rft_id=info%3Apmid%2F40577294&rft.externalDocID=11054071 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=2162-237X&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=2162-237X&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=2162-237X&client=summon |