Research map / 3D vision / NeRF and radiance fields
Acquiring Dynamic Light Fields through Coded Aperture Camera
Kohei Sakai, Keita Takahashi, Toshiaki Fujii, Hajime Nagahara
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| train on dynamic light-field dataset with pseudo motions instead of static dataset (2-D vs 2-S, two consecutive images) (ablation) | 2-frame reconstruction CNN · Synthetic dynamic scene (5x5 views) | PSNR | Improved |
| use three consecutive images in V-shape observation instead of two consecutive images (3-D vs 2-D) (ablation) | Dynamic light-field CNN · Synthetic dynamic scene (5x5 views) | PSNR | Improved |
| add a second observed image (2-S) to single-image reconstruction when both trained on static data (ablation) | Inagaki et al. static-trained network · Synthetic dynamic scene | PSNR | Hurt |
| V-shape 3-D method versus Reference using two images observed at the same time instant | 3-D (V-shape) · Synthetic dynamic scene | PSNR | No clear effect |
| use V-shape 3-D method instead of 2-S on a physical coded aperture camera | Physical LCoS coded aperture camera (5x5 views) · Real turntable scene | qualitative reconstruction quality | Improved |
Closest papers by meaning
- Acquiring a Dynamic Light Field Through a Single-Shot Coded Image (CVPR 2022 · similarity 0.96)
- Learning to Capture Light Fields through a Coded Aperture Camera (ECCV 2018 · similarity 0.93)
- Time-Efficient Light-Field Acquisition Using Coded Aperture and Events (CVPR 2024 · similarity 0.90)
- Deep Spatial-angular Regularization for Compressive Light Field Reconstruction over Coded Apertures (ECCV 2020 · similarity 0.89)
- Event Fields: Capturing Light Fields at High Speed, Resolution, and Dynamic Range (CVPR 2025 · similarity 0.86)
- Real-time Acquisition and Reconstruction of Dynamic Volumes with Neural Structured Illumination (CVPR 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
