Research map / Image generation / Image restoration
Probabilistic-Based Feature Embedding of 4-D Light Fields for Compressive Imaging and Denoising
Xianqiang Lyu, Junhui Hou
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 |
|---|---|---|---|
| add probabilistic modeling (vs template network) (ablation) | CR-Net · HCI+Inria S=2 | PSNR: 42.12 -> 42.49 | Improved |
| add PFE instead of 3-D convolution (ablation) | CR-Net · HCI+Inria S=2 | PSNR: 40.42 -> 42.49 | Improved |
| add PFE instead of SAS layers (ablation) | CR-Net · HCI+Inria S=2 | PSNR: 41.11 -> 42.49 | Improved |
| share projection layer weights across stages (ablation) | CR-Net · HCI+Inria S=2 | PSNR: 41.67 -> 42.49 | Improved |
| increase stages T from 2 to 6 (ablation) | CR-Net · HCI+Inria S=2 | PSNR: 41.53 -> 42.49 | Improved |
| increase feature embedding units J from 8 to 10 (ablation) | CR-Net · HCI+Inria S=2 | PSNR: 42.49 -> 41.91 | Hurt |
| increase feature embedding units J from 6 to 8 (ablation) | CR-Net · HCI+Inria S=2 | PSNR: 41.55 -> 42.49 | Improved |
| add noise suppression module (ablation) | DN-Net · HCI+Inria sigma=20 | PSNR: 40.04 -> 40.50 | Improved |
| add PFE instead of SAS layers (ablation) | DN-Net · HCI+Inria sigma=20 | PSNR: 38.82 -> 40.50 | Improved |
| add PFE instead of 3-D convolution (ablation) | DN-Net · HCI+Inria sigma=20 | PSNR: 36.82 -> 40.50 | Improved |
| add probabilistic modeling (vs template network) (ablation) | DN-Net · HCI+Inria sigma=20 | PSNR: 40.27 -> 40.50 | Improved |
| increase iterative stages T from 1 to 4 (ablation) | DN-Net · HCI+Inria sigma=20 | PSNR: 40.01 -> 40.76 | Improved |
Closest papers by meaning
- Deep Spatial-Angular Regularization for Light Field Imaging, Denoising, and Super-Resolution (TPAMI 2021 · similarity 0.90)
- Deep Spatial-angular Regularization for Compressive Light Field Reconstruction over Coded Apertures (ECCV 2020 · similarity 0.89)
- High-Dimensional Dense Residual Convolutional Neural Network for Light Field Reconstruction (TPAMI 2019 · similarity 0.87)
- From Pixels to Views: Learning Angular-Aware and Physics-Consistent Representations for Light Field Microscopy (NeurIPS 2025 · similarity 0.86)
- NeLF-Pro: Neural Light Field Probes for Multi-Scale Novel View Synthesis (CVPR 2024 · similarity 0.85)
- V2V3D: View-to-View Denoised 3D Reconstruction for Light Field Microscopy (CVPR 2025 · similarity 0.84)
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.
