Deep Stereo Using Adaptive Thin Volume Representation With Uncertainty Awareness
We present Uncertainty-aware Cascaded Stereo Network (UCS-Net) for 3D reconstruction from multiple RGB images. Multi-view stereo (MVS) aims to reconstruct fine-grained scene geometry from multi-view images. Previous learning-based MVS methods estimate per-view depth using plane sweep volumes (PSVs)...
Gespeichert in:
| Veröffentlicht in: | Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) S. 2521 - 2531 |
|---|---|
| Hauptverfasser: | , , , , , , |
| Format: | Tagungsbericht |
| Sprache: | Englisch |
| Veröffentlicht: |
IEEE
01.06.2020
|
| Schlagworte: | |
| ISSN: | 1063-6919 |
| Online-Zugang: | Volltext |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| Abstract | We present Uncertainty-aware Cascaded Stereo Network (UCS-Net) for 3D reconstruction from multiple RGB images. Multi-view stereo (MVS) aims to reconstruct fine-grained scene geometry from multi-view images. Previous learning-based MVS methods estimate per-view depth using plane sweep volumes (PSVs) with a fixed depth hypothesis at each plane; this requires densely sampled planes for high accuracy, which is impractical for high-resolution depth because of limited memory. In contrast, we propose adaptive thin volumes (ATVs); in an ATV, the depth hypothesis of each plane is spatially varying, which adapts to the uncertainties of previous per-pixel depth predictions. Our UCS-Net has three stages: the first stage processes a small PSV to predict low-resolution depth; two ATVs are then used in the following stages to refine the depth with higher resolution and higher accuracy. Our ATV consists of only a small number of planes with low memory and computation costs; yet, it efficiently partitions local depth ranges within learned small uncertainty intervals. We propose to use variance-based uncertainty estimates to adaptively construct ATVs; this differentiable process leads to reasonable and fine-grained spatial partitioning. Our multi-stage framework progressively sub-divides the vast scene space with increasing depth resolution and precision, which enables reconstruction with high completeness and accuracy in a coarse-to-fine fashion. We demonstrate that our method achieves superior performance compared with other learning-based MVS methods on various challenging datasets. |
|---|---|
| AbstractList | We present Uncertainty-aware Cascaded Stereo Network (UCS-Net) for 3D reconstruction from multiple RGB images. Multi-view stereo (MVS) aims to reconstruct fine-grained scene geometry from multi-view images. Previous learning-based MVS methods estimate per-view depth using plane sweep volumes (PSVs) with a fixed depth hypothesis at each plane; this requires densely sampled planes for high accuracy, which is impractical for high-resolution depth because of limited memory. In contrast, we propose adaptive thin volumes (ATVs); in an ATV, the depth hypothesis of each plane is spatially varying, which adapts to the uncertainties of previous per-pixel depth predictions. Our UCS-Net has three stages: the first stage processes a small PSV to predict low-resolution depth; two ATVs are then used in the following stages to refine the depth with higher resolution and higher accuracy. Our ATV consists of only a small number of planes with low memory and computation costs; yet, it efficiently partitions local depth ranges within learned small uncertainty intervals. We propose to use variance-based uncertainty estimates to adaptively construct ATVs; this differentiable process leads to reasonable and fine-grained spatial partitioning. Our multi-stage framework progressively sub-divides the vast scene space with increasing depth resolution and precision, which enables reconstruction with high completeness and accuracy in a coarse-to-fine fashion. We demonstrate that our method achieves superior performance compared with other learning-based MVS methods on various challenging datasets. |
| Author | Xu, Zexiang Zhu, Shilin Li, Zhuwen Su, Hao Ramamoorthi, Ravi Li, Li Erran Cheng, Shuo |
| Author_xml | – sequence: 1 givenname: Shuo surname: Cheng fullname: Cheng, Shuo organization: University of California, San Diego – sequence: 2 givenname: Zexiang surname: Xu fullname: Xu, Zexiang organization: University of California, San Diego – sequence: 3 givenname: Shilin surname: Zhu fullname: Zhu, Shilin organization: University of California, San Diego – sequence: 4 givenname: Zhuwen surname: Li fullname: Li, Zhuwen organization: Nuro Inc – sequence: 5 givenname: Li Erran surname: Li fullname: Li, Li Erran organization: Scale AI; Columbia University – sequence: 6 givenname: Ravi surname: Ramamoorthi fullname: Ramamoorthi, Ravi organization: University of California, San Diego – sequence: 7 givenname: Hao surname: Su fullname: Su, Hao organization: University of California, San Diego |
| BookMark | eNotjNFOwjAUQKvRRES-QB_6A8Pb3rVrHwmKmpBIcOAjGdut1EC3rFXD30uiDyfnPJ1rdhHaQIzdCRgLAfZ-ul4sc6kBxhIkjAFOfcZGtjCikCeENuqcDQRozLQV9oqNYvwEAJRCaGsGbPFA1PG3RD21fBV9-OCTpuqS_yZe7nzg63b_dSC-pK6nSCFVybeBv_u046tQU58qH9KRT36qngLFeMMuXbWPNPr3kJWzx3L6nM1fn16mk3nmJWDKnFOFgcblEpotOcqN0Y2pkQgBwRpqsMa6diC22jpUW4UFKiCjcmWcxSG7_dt6Itp0vT9U_XFjhdKYS_wFnwdSWA |
| CODEN | IEEPAD |
| ContentType | Conference Proceeding |
| DBID | 6IE 6IH CBEJK RIE RIO |
| DOI | 10.1109/CVPR42600.2020.00260 |
| DatabaseName | IEEE Electronic Library (IEL) Conference Proceedings IEEE Proceedings Order Plan (POP) 1998-present by volume IEEE Xplore All Conference Proceedings IEEE Electronic Library (IEL) IEEE Proceedings Order Plans (POP) 1998-present |
| DatabaseTitleList | |
| Database_xml | – sequence: 1 dbid: RIE name: IEEE Electronic Library (IEL) url: https://ieeexplore.ieee.org/ sourceTypes: Publisher |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Applied Sciences |
| EISBN | 9781728171685 1728171687 |
| EISSN | 1063-6919 |
| EndPage | 2531 |
| ExternalDocumentID | 9156342 |
| Genre | orig-research |
| GroupedDBID | 6IE 6IH 6IL 6IN AAWTH ABLEC ADZIZ ALMA_UNASSIGNED_HOLDINGS BEFXN BFFAM BGNUA BKEBE BPEOZ CBEJK CHZPO IEGSK IJVOP OCL RIE RIL RIO |
| ID | FETCH-LOGICAL-i203t-ff5780df420dbefe4886d8c3ee303098ed3c3ccf01b69f35b537350e85458f93 |
| IEDL.DBID | RIE |
| ISICitedReferencesCount | 290 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=000620679502079&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| IngestDate | Wed Aug 27 02:30:35 EDT 2025 |
| IsPeerReviewed | false |
| IsScholarly | true |
| Language | English |
| LinkModel | DirectLink |
| MergedId | FETCHMERGED-LOGICAL-i203t-ff5780df420dbefe4886d8c3ee303098ed3c3ccf01b69f35b537350e85458f93 |
| PageCount | 11 |
| ParticipantIDs | ieee_primary_9156342 |
| PublicationCentury | 2000 |
| PublicationDate | 2020-Jun |
| PublicationDateYYYYMMDD | 2020-06-01 |
| PublicationDate_xml | – month: 06 year: 2020 text: 2020-Jun |
| PublicationDecade | 2020 |
| PublicationTitle | Proceedings (IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Online) |
| PublicationTitleAbbrev | CVPR |
| PublicationYear | 2020 |
| Publisher | IEEE |
| Publisher_xml | – name: IEEE |
| SSID | ssj0003211698 |
| Score | 2.6122873 |
| Snippet | We present Uncertainty-aware Cascaded Stereo Network (UCS-Net) for 3D reconstruction from multiple RGB images. Multi-view stereo (MVS) aims to reconstruct... |
| SourceID | ieee |
| SourceType | Publisher |
| StartPage | 2521 |
| SubjectTerms | Feature extraction Geometry Image reconstruction Image resolution Memory management Three-dimensional displays Uncertainty |
| Title | Deep Stereo Using Adaptive Thin Volume Representation With Uncertainty Awareness |
| URI | https://ieeexplore.ieee.org/document/9156342 |
| WOSCitedRecordID | wos000620679502079&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 | |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1LTwIxEG6QePDkA4zv9ODRldIpbfdIUOLBkA0iciO77TRyAQKLxn9vWzbowYu3poc2mclMOzPfN0PILRdOc5VjAlJiIkC6pFBgE3RC-d-zLfJYPR8_q8FATyZpViN3Oy4MIkbwGd6HZazl24XZhFRZK_XBBgjvcPeUkluu1i6fAj6Skamu2HFtlrZ642wY-6_7KJAHABcPfSh_zVCJT0j_8H-XH5HmDxePZrtX5pjUcH5CDqvPI61Mc90g2QPikr54MeGCRiAA7dp8GbwZDcM56Tj6ITqM0NeKcTSnb7Pynb76IyIyoPyi3c9AD_P-r0lG_cdR7ympxiUkM86gTJzz1sesE5zZAh1605RWG0CEUEfRaMGAMY61C5k66BQdUNBhqEPtzKVwSurzxRzPCAXtjCwKLbhhAoVJmfNmLHLjwDFU_Jw0gnymy21DjGklmou_ty_JQVDAFl91RerlaoPXZN98lLP16iZq8Rv_ip7d |
| linkProvider | IEEE |
| linkToHtml | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwlV1NTwIxEG0ImugJFYzf9uDRlbLT3e0eCUowItkgIjey204jl4XAovHf25YNevDiremhTWYy087MezOE3PhcCz9K0YMwRI9DqL0sAuWh5pH5PassddXzcT8aDMRkEicVcrvlwiCiA5_hnV26Wr6ay7VNlTVjE2wANw53J-DcZxu21jajAiaWCWNR8uNaLG52xsnQdWA3caBvIVy-7UT5a4qKe0S6tf9df0AaP2w8mmzfmUNSwfyI1MrvIy2Nc1UnyT3igr4YQeGcOigAbat0Yf0ZteM56dh5Ijp04NeSc5TTt1nxTl_NEQ4bUHzR9qcliBkP2CCj7sOo0_PKgQnezGdQeFob-2NKG9moDDUa4wyVkIAItpIiUIEEKTVrZWGsIcgCiCBgKGz1TMdwTKr5PMcTQkFoGWaZ4L5kHLmMmTaGzFOpQTOM_FNSt_KZLjYtMaalaM7-3r4me73Rc3_afxw8nZN9q4wN2uqCVIvlGi_JrvwoZqvlldPoN3HToiQ |
| 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%3Abook&rft.genre=proceeding&rft.title=Proceedings+%28IEEE+Computer+Society+Conference+on+Computer+Vision+and+Pattern+Recognition.+Online%29&rft.atitle=Deep+Stereo+Using+Adaptive+Thin+Volume+Representation+With+Uncertainty+Awareness&rft.au=Cheng%2C+Shuo&rft.au=Xu%2C+Zexiang&rft.au=Zhu%2C+Shilin&rft.au=Li%2C+Zhuwen&rft.date=2020-06-01&rft.pub=IEEE&rft.eissn=1063-6919&rft.spage=2521&rft.epage=2531&rft_id=info:doi/10.1109%2FCVPR42600.2020.00260&rft.externalDocID=9156342 |