File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1145/3355089.3356526
- Scopus: eid_2-s2.0-85078922577
- WOS: WOS:000498397300068
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: Learned large field-of-view imaging with thin-plate optics
Title | Learned large field-of-view imaging with thin-plate optics |
---|---|
Authors | |
Keywords | Computational camera Deep network Image deblurring Thin optics |
Issue Date | 2019 |
Citation | ACM Transactions on Graphics, 2019, v. 38, n. 6, article no. 3356526 How to Cite? |
Abstract | Typical camera optics consist of a system of individual elements that are designed to compensate for the aberrations of a single lens. Recent computational cameras shift some of this correction task from the optics to post-capture processing, reducing the imaging optics to only a few optical elements. However, these systems only achieve reasonable image quality by limiting the field of view (FOV) to a few degrees - effectively ignoring severe off-axis aberrations with blur sizes of multiple hundred pixels. In this paper, we propose a lens design and learned reconstruction architecture that lift this limitation and provide an order of magnitude increase in field of view using only a single thin-plate lens element. Specifically, we design a lens to produce spatially shift-invariant point spread functions, over the full FOV, that are tailored to the proposed reconstruction architecture. We achieve this with a mixture PSF, consisting of a peak and and a low-pass component, which provides residual contrast instead of a small spot size as in traditional lens designs. To perform the reconstruction, we train a deep network on captured data from a display lab setup, eliminating the need for manual acquisition of training data in the field. We assess the proposed method in simulation and experimentally with a prototype camera system. We compare our system against existing single-element designs, including an aspherical lens and a pinhole, and we compare against a complex multielement lens, validating high-quality large field-of-view (i.e. 53°) imaging performance using only a single thin-plate element. |
Persistent Identifier | http://hdl.handle.net/10722/315316 |
ISSN | 2023 Impact Factor: 7.8 2023 SCImago Journal Rankings: 7.766 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Peng, Yifan | - |
dc.contributor.author | Sun, Qilin | - |
dc.contributor.author | Dun, Xiong | - |
dc.contributor.author | Wetzstein, Gordon | - |
dc.contributor.author | Heidrich, Wolfgang | - |
dc.contributor.author | Heide, Felix | - |
dc.date.accessioned | 2022-08-05T10:18:27Z | - |
dc.date.available | 2022-08-05T10:18:27Z | - |
dc.date.issued | 2019 | - |
dc.identifier.citation | ACM Transactions on Graphics, 2019, v. 38, n. 6, article no. 3356526 | - |
dc.identifier.issn | 0730-0301 | - |
dc.identifier.uri | http://hdl.handle.net/10722/315316 | - |
dc.description.abstract | Typical camera optics consist of a system of individual elements that are designed to compensate for the aberrations of a single lens. Recent computational cameras shift some of this correction task from the optics to post-capture processing, reducing the imaging optics to only a few optical elements. However, these systems only achieve reasonable image quality by limiting the field of view (FOV) to a few degrees - effectively ignoring severe off-axis aberrations with blur sizes of multiple hundred pixels. In this paper, we propose a lens design and learned reconstruction architecture that lift this limitation and provide an order of magnitude increase in field of view using only a single thin-plate lens element. Specifically, we design a lens to produce spatially shift-invariant point spread functions, over the full FOV, that are tailored to the proposed reconstruction architecture. We achieve this with a mixture PSF, consisting of a peak and and a low-pass component, which provides residual contrast instead of a small spot size as in traditional lens designs. To perform the reconstruction, we train a deep network on captured data from a display lab setup, eliminating the need for manual acquisition of training data in the field. We assess the proposed method in simulation and experimentally with a prototype camera system. We compare our system against existing single-element designs, including an aspherical lens and a pinhole, and we compare against a complex multielement lens, validating high-quality large field-of-view (i.e. 53°) imaging performance using only a single thin-plate element. | - |
dc.language | eng | - |
dc.relation.ispartof | ACM Transactions on Graphics | - |
dc.subject | Computational camera | - |
dc.subject | Deep network | - |
dc.subject | Image deblurring | - |
dc.subject | Thin optics | - |
dc.title | Learned large field-of-view imaging with thin-plate optics | - |
dc.type | Article | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1145/3355089.3356526 | - |
dc.identifier.scopus | eid_2-s2.0-85078922577 | - |
dc.identifier.volume | 38 | - |
dc.identifier.issue | 6 | - |
dc.identifier.spage | article no. 3356526 | - |
dc.identifier.epage | article no. 3356526 | - |
dc.identifier.eissn | 1557-7368 | - |
dc.identifier.isi | WOS:000498397300068 | - |