Acquiring a Dynamic Light Field through a Single-Shot Coded Image

Mizuno, Ryoya; Takahashi, Keita; Yoshida, Masahito; Tsutake, Chihiro; Fujii, Toshiaki; Nagahara, Hajime

doi:10.1109/cvpr52688.2022.01921

Cited by 6 publications

(21 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our method achieved comparable quality to Mizuno et al's [24] for these discretized viewpoints. (re-trained) [24] Ours Ours (center-only) For comparison, we tested two other methods for continuous light-field reconstruction. The first is MINE [51], a stateof-the-art single-view view synthesis method without test time training.…”

Section: B Continuous Light-field Reconstructionsupporting

confidence: 58%

“…Using joint aperture-exposure coding for image acquisition, our method achieved a reconstruction quality comparable to Mizuno et al's [24] for the discretized 5 × 5 viewpoints. The quantitative scores for Mizuno et al's [24] were slightly better than those for ours; this is understandable, because Mizuno et al's [24] devoted all the resources to the discrete 5 × 5 views without considering other viewpoints. Meanwhile, ours (no-coding) and ours (center-only) produced poor results, showing the difficulty of light field reconstruction from a single image without camera-side coding.…”

Section: A Compressive Light-field Acquisitionsupporting

confidence: 52%

“…Note that our method can reconstruct continuous viewpoints, whereas the others [19], [21], [24] can reconstruct only discretized 5 × 5 views. In other words, these methods [19], [21], [24] devoted all resources to reconstructing individual 5 × 5 views rather than a continuous 3-D representation of the target scene.…”

Section: A Compressive Light-field Acquisitionmentioning

confidence: 99%

“…Joint aperture-exposure coding (Mizuno et al's [24]) significantly outperformed aperture coding (Inagaki et al's [19] and Guo et al's [21]). Using joint aperture-exposure coding for image acquisition, our method achieved a reconstruction quality comparable to Mizuno et al's [24] for the discretized 5 × 5 viewpoints. The quantitative scores for Mizuno et al's [24] were slightly better than those for ours; this is understandable, because Mizuno et al's [24] devoted all the resources to the discrete 5 × 5 views without considering other viewpoints.…”

Section: A Compressive Light-field Acquisitionmentioning

confidence: 99%

“…Note that our method can reconstruct continuous light fields, whereas others [19], [21], [24] can obtain only discretized 5 × 5 views. Our method achieved comparable quality to Mizuno et al's [24] for these discretized viewpoints. (re-trained) [24] Ours Ours (center-only) For comparison, we tested two other methods for continuous light-field reconstruction.…”

Section: B Continuous Light-field Reconstructionmentioning

confidence: 99%

See 4 more Smart Citations

Reconstructing Continuous Light Field From Single Coded Image

Ishikawa,

Takahashi,

Tsutake

et al. 2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

We propose a method for reconstructing a continuous light field of a target scene from a single observed image. Our method takes the best of two worlds: joint aperture-exposure coding for compressive light-field acquisition, and a neural radiance field (NeRF) for view synthesis. Joint aperture-exposure coding implemented in a camera enables effective embedding of 3-D scene information into an observed image, but in previous works, it was used only for reconstructing discretized light-field views. NeRF-based neural rendering enables high quality view synthesis of a 3-D scene from continuous viewpoints, but when only a single image is given as the input, it struggles to achieve satisfactory quality. Our method integrates these two techniques into an efficient and end-to-end trainable pipeline. Trained on a wide variety of scenes, our method can reconstruct continuous light fields accurately and efficiently without any test time optimization. To our knowledge, this is the first work to bridge two worlds: camera design for efficiently acquiring 3-D information and neural rendering.

show abstract

Section: B Continuous Light-field Reconstructionsupporting

confidence: 58%

Section: A Compressive Light-field Acquisitionsupporting

confidence: 52%

Section: A Compressive Light-field Acquisitionmentioning

confidence: 99%

Section: A Compressive Light-field Acquisitionmentioning

confidence: 99%

Section: B Continuous Light-field Reconstructionmentioning

confidence: 99%

See 3 more Smart Citations

Reconstructing Continuous Light Field From Single Coded Image

Ishikawa,

Takahashi,

Tsutake

et al. 2023

IEEE Access

Self Cite

View full text Add to dashboard Cite

show abstract

Pixel-wise exposure control for single-shot HDR imaging: A joint optimization approach

Liang,

Huang,

et al. 2024

APL Photonics

View full text Add to dashboard Cite

Dynamic range is one of the primary limitations that restricts digital image sensors from acquiring more visual information. Current high dynamic range (HDR) imaging techniques entail a trade-off between dynamic range and visual fidelity. In this work, we propose a HDR imaging method, termed PE-HDR, to achieve both a wide dynamic range and high visual fidelity without additional complex post-processing algorithms. Instead of merging a bracketed exposure sequence, the PE-HDR captures HDR images in a single shot using optical coded pixel-wise exposure control, enabling cost-effective and flexible HDR imaging. By incorporating a differentiable optical encoder and a neural network decoder, we jointly optimize the imaging pipeline from light irradiance to digital image signals, thereby refining the pixel-wise exposure control strategy and improving image fidelity. Both simulations and experiments demonstrate that the proposed method achieves a dynamic range of up to 120 dB and an excellent visual fidelity with spatial resolution of up to 2560 × 1600 pixels.

show abstract