Research map / 3D vision / Depth and stereo
Learning Fused Pixel and Feature-Based View Reconstructions for Light Fields
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Combine pixel-wise and feature-based reconstructions with a learned soft mask (ablation) | FPFR · Synthetic light-field test scenes | PSNR | Helped |
| Use feature-based reconstruction instead of perceptual-loss supervision for pixel reconstruction (ablation) | FPFR · Synthetic light-field test scenes | PSNR: about 0.7dB | Helped |
| Use multi-scale rather than single-scale feature reconstruction (ablation) | FeatRNet · Synthetic light-field test scenes | PSNR: about 0.5dB | Helped |
| Add disparity-dependent interpolation for occlusion handling and learned inpainting for disoccluded areas (ablation) | PixRNet · Light-field view synthesis | PSNR | Helped |
| Use warped VGG features as inputs to reconstruct target-view features, then infer the view (ablation) | FeatRNet · Highly textured regions and object contours | PSNR | Helped |
| Apply test-time rotation and flipping and average the eight reconstructions (ablation) | FPFR · Synthetic light-field test scenes | PSNR | Helped |
| Use the trained interpolation model for extrapolation without additional training | FPFR · Synthetic light-field extrapolation | PSNR | Helped |
Closest papers by meaning
- Learning Optical Flow From Still Images (CVPR 2021 · similarity 0.88)
- Attention-based Multi-Level Fusion Network for Light Field Depth Estimation (AAAI 2021 · similarity 0.88)
- Attention-Based View Selection Networks for Light-Field Disparity Estimation (AAAI 2020 · similarity 0.88)
- Light Field Neural Rendering (CVPR 2022 · similarity 0.88)
- NeLF-Pro: Neural Light Field Probes for Multi-Scale Novel View Synthesis (CVPR 2024 · similarity 0.86)
- Flow-NeRF: Joint Learning of Geometry, Poses, and Dense Flow within Unified Neural Representations (CVPR 2025 · 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.