DC Field | Value | Language |
---|---|---|
dc.contributor.author | Khan, Numair | ko |
dc.contributor.author | Kim, Min H. | ko |
dc.contributor.author | Tompkin, James | ko |
dc.date.accessioned | 2024-08-29T08:00:06Z | - |
dc.date.available | 2024-08-29T08:00:06Z | - |
dc.date.created | 2024-05-18 | - |
dc.date.issued | 2024-07 | - |
dc.identifier.citation | INTERNATIONAL JOURNAL OF COMPUTER VISION, v.132, no.7, pp.2639 - 2673 | - |
dc.identifier.issn | 0920-5691 | - |
dc.identifier.uri | http://hdl.handle.net/10203/322456 | - |
dc.description.abstract | Depth estimation tries to obtain 3D scene geometry from low-dimensional data like 2D images. This is a vital operation in computer vision and any general solution must preserve all depth information of potential relevance to support higher-level tasks. For scenes with well-defined depth, this work shows that multi-view edges can encode all relevant information-that multi-view edges are complete. For this, we follow Elder's complementary work on the completeness of 2D edges for image reconstruction. We deploy an image-space geometric representation: an encoding of multi-view scene edges as constraints and a diffusion reconstruction method for inverting this code into depth maps. Due to inaccurate constraints, diffusion-based methods have previously underperformed against deep learning methods; however, we will reassess the value of diffusion-based methods and show their competitiveness without requiring training data. To begin, we work with structured light fields and epipolar plane images (EPIs). EPIs present high-gradient edges in the angular domain: with correct processing, EPIs provide depth constraints with accurate occlusion boundaries and view consistency. Then, we present a differentiable representation form that allows the constraints and the diffusion reconstruction to be optimized in an unsupervised way via a multi-view reconstruction loss. This is based around point splatting via radiative transport, and extends to unstructured multi-view images. We evaluate our reconstructions for accuracy, occlusion handling, view consistency, and sparsity to show that they retain the geometric information required for higher-level tasks. | - |
dc.language | English | - |
dc.publisher | SPRINGER | - |
dc.title | Are Multi-view Edges Incomplete for Depth Estimation? | - |
dc.type | Article | - |
dc.identifier.wosid | 001159361200001 | - |
dc.identifier.scopusid | 2-s2.0-85184885550 | - |
dc.type.rims | ART | - |
dc.citation.volume | 132 | - |
dc.citation.issue | 7 | - |
dc.citation.beginningpage | 2639 | - |
dc.citation.endingpage | 2673 | - |
dc.citation.publicationname | INTERNATIONAL JOURNAL OF COMPUTER VISION | - |
dc.identifier.doi | 10.1007/s11263-023-01890-y | - |
dc.contributor.localauthor | Kim, Min H. | - |
dc.contributor.nonIdAuthor | Khan, Numair | - |
dc.contributor.nonIdAuthor | Tompkin, James | - |
dc.description.isOpenAccess | N | - |
dc.type.journalArticle | Article | - |
dc.subject.keywordAuthor | Diffusion | - |
dc.subject.keywordAuthor | Light fields | - |
dc.subject.keywordAuthor | Multi-view reconstruction | - |
dc.subject.keywordAuthor | Edges | - |
dc.subject.keywordAuthor | Depth reconstruction | - |
dc.subject.keywordPlus | FIELD | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.