Research map / 3D vision / NeRF and radiance fields
Learning Dynamic Interpolation for Extremely Sparse Light Fields With Wide Baselines
Mantang Guo, Jing Jin, Hui Liu, 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 content embedding (ablation) | Base model · Inria Sparse LF (7 scenes) (ablation) | PSNR: 29.05 -> 29.83 | Improved |
| add geometry-based spatial refinement (ablation) | Base + content embedding · Inria Sparse LF (7 scenes) (ablation) | PSNR: 29.83 -> 30.14 | Improved |
| add disparity-oriented loss (ablation) | Base + content embedding + refinement · Inria Sparse LF (7 scenes) (ablation) | PSNR: 30.14 -> 30.33 | Improved |
| add content embedding (ablation) | Base model · Inria Sparse LF (7 scenes) (ablation) | SSIM: 0.901 -> 0.915 | Improved |
| add geometry-based spatial refinement (ablation) | Base + content embedding · Inria Sparse LF (7 scenes) (ablation) | SSIM: 0.915 -> 0.920 | Improved |
| add disparity-oriented loss (ablation) | Base + content embedding + refinement · Inria Sparse LF (7 scenes) (ablation) | SSIM: 0.920 -> 0.919 | No clear effect |
| use dynamic interpolation weights instead of disparity-based warping | Baseline · Inria Sparse LF (7 scenes) average | PSNR: 25.77 -> 30.33 | Improved |
| use RAFT instead of PWCNet for disparity estimation (ablation) | Ours · Inria Sparse LF (7 scenes) average | PSNR: 28.53 -> 30.33 | Improved |
| use RAFT instead of PWCNet for disparity estimation (ablation) | Ours · Inria Sparse LF (7 scenes) average | SSIM: 0.893 -> 0.919 | Improved |
| use dynamic interpolation instead of warping | Baseline · Inria Sparse LF (7 scenes) | Running time (s per view): 0.65 -> 2.78 | Hurt |
Closest papers by meaning
- Geo-NI: Geometry-Aware Neural Interpolation for Light Field Rendering (TPAMI 2025 · similarity 0.90)
- Deep Coarse-to-Fine Dense Light Field Reconstruction With Flexible Sampling and Geometry-Aware Fusion (TPAMI 2020 · similarity 0.90)
- Light Field Neural Rendering (CVPR 2022 · similarity 0.89)
- Fast Light Field Reconstruction With Deep Coarse-To-Fine Modeling of Spatial-Angular Clues (ECCV 2018 · similarity 0.89)
- Revisiting Light Field Rendering with Deep Anti-Aliasing Neural Network (TPAMI 2021 · similarity 0.89)
- Learning Fused Pixel and Feature-Based View Reconstructions for Light Fields (CVPR 2020 · similarity 0.89)
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.
