Deep Unfolding-Aided Parameter Tuning for Plug-and-Play-Based Video Snapshot Compressive Imaging
Snapshot compressive imaging (SCI) captures high-dimensional data efficiently by compressing it into two-dimensional observations and reconstructing high-dimensional data from two-dimensional observations with various algorithms. The plug-and-play (PnP) method is a promising approach for the video S...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10870238/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1823857118647681024 |
|---|---|
| author | Takashi Matsuda Ryo Hayakawa Youji Iiguni |
| author_facet | Takashi Matsuda Ryo Hayakawa Youji Iiguni |
| author_sort | Takashi Matsuda |
| collection | DOAJ |
| description | Snapshot compressive imaging (SCI) captures high-dimensional data efficiently by compressing it into two-dimensional observations and reconstructing high-dimensional data from two-dimensional observations with various algorithms. The plug-and-play (PnP) method is a promising approach for the video SCI reconstruction because it can leverage both observation models and denoising methods for videos. Since the reconstruction accuracy significantly depends on the choice of noise level parameters, this paper proposes a deep unfolding-based method for tuning these parameters in PnP-based video SCI. For the training of the parameters, we prepare training data from the densely annotated video segmentation dataset, reparametrize the noise level parameters, and apply the checkpointing technique to reduce the required memory. Simulation results show that the trained noise level parameters via the proposed approach exhibit a non-monotonic pattern, which is different from the assumptions in the conventional convergence analyses of PnP-based algorithms. These findings provide new insights into both the application of deep unfolding and the theoretical basis of PnP algorithms. |
| format | Article |
| id | doaj-art-2d5b7f1bf0ff43d19b96f8a3a10e18f8 |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-2d5b7f1bf0ff43d19b96f8a3a10e18f82025-02-12T00:02:52ZengIEEEIEEE Access2169-35362025-01-0113248672487910.1109/ACCESS.2025.353849910870238Deep Unfolding-Aided Parameter Tuning for Plug-and-Play-Based Video Snapshot Compressive ImagingTakashi Matsuda0Ryo Hayakawa1https://orcid.org/0000-0003-4510-6258Youji Iiguni2Graduate School of Engineering Science, Osaka University, Toyonaka, Osaka, JapanInstitute of Engineering, Tokyo University of Agriculture and Technology, Koganei, Tokyo, JapanGraduate School of Engineering Science, Osaka University, Toyonaka, Osaka, JapanSnapshot compressive imaging (SCI) captures high-dimensional data efficiently by compressing it into two-dimensional observations and reconstructing high-dimensional data from two-dimensional observations with various algorithms. The plug-and-play (PnP) method is a promising approach for the video SCI reconstruction because it can leverage both observation models and denoising methods for videos. Since the reconstruction accuracy significantly depends on the choice of noise level parameters, this paper proposes a deep unfolding-based method for tuning these parameters in PnP-based video SCI. For the training of the parameters, we prepare training data from the densely annotated video segmentation dataset, reparametrize the noise level parameters, and apply the checkpointing technique to reduce the required memory. Simulation results show that the trained noise level parameters via the proposed approach exhibit a non-monotonic pattern, which is different from the assumptions in the conventional convergence analyses of PnP-based algorithms. These findings provide new insights into both the application of deep unfolding and the theoretical basis of PnP algorithms.https://ieeexplore.ieee.org/document/10870238/Deep unfoldingparameter tuningplug and playsnapshot compressive imaging |
| spellingShingle | Takashi Matsuda Ryo Hayakawa Youji Iiguni Deep Unfolding-Aided Parameter Tuning for Plug-and-Play-Based Video Snapshot Compressive Imaging IEEE Access Deep unfolding parameter tuning plug and play snapshot compressive imaging |
| title | Deep Unfolding-Aided Parameter Tuning for Plug-and-Play-Based Video Snapshot Compressive Imaging |
| title_full | Deep Unfolding-Aided Parameter Tuning for Plug-and-Play-Based Video Snapshot Compressive Imaging |
| title_fullStr | Deep Unfolding-Aided Parameter Tuning for Plug-and-Play-Based Video Snapshot Compressive Imaging |
| title_full_unstemmed | Deep Unfolding-Aided Parameter Tuning for Plug-and-Play-Based Video Snapshot Compressive Imaging |
| title_short | Deep Unfolding-Aided Parameter Tuning for Plug-and-Play-Based Video Snapshot Compressive Imaging |
| title_sort | deep unfolding aided parameter tuning for plug and play based video snapshot compressive imaging |
| topic | Deep unfolding parameter tuning plug and play snapshot compressive imaging |
| url | https://ieeexplore.ieee.org/document/10870238/ |
| work_keys_str_mv | AT takashimatsuda deepunfoldingaidedparametertuningforplugandplaybasedvideosnapshotcompressiveimaging AT ryohayakawa deepunfoldingaidedparametertuningforplugandplaybasedvideosnapshotcompressiveimaging AT youjiiiguni deepunfoldingaidedparametertuningforplugandplaybasedvideosnapshotcompressiveimaging |