Research map / 3D vision / Depth and stereo
BokehMe: When Neural Rendering Meets Classical Rendering
Juewen Peng, Zhiguo Cao, Xianrui Luo, Hao Lu, Ke Xian, Jianming Zhang
Check 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 |
|---|---|---|---|
| combine the classical and neural renderers using an error map | BokehMe · BLB Levels 1–5 | PSNR: 43.30 vs 41.32; 42.21 vs 40.16; 41.02 vs 39.21; 39.78 vs 38.01; 38.80 vs 37.20 | Improved |
| train with disparity augmentation | BokehMe · BLB with corrupted disparity maps | PSNR | Improved |
| add IUNet iterative upsampling instead of bilinear upsampling (ablation) | BokehMe neural renderer · BLB Level 3 | PSNR: 37.30 vs 23.23 | Mixed |
| add clipping of the signed defocus map during iterative upsampling (ablation) | BokehMe neural renderer · BLB Level 3 | PSNR: 23.23 -> 38.55 | Improved |
| add fusion that uses bilinearly upsampled bokeh in clipped regions (ablation) | BokehMe neural renderer · BLB Level 3 | PSNR: +0.64 dB | Improved |
| use a dilated rather than signed defocus map for the fusion mask (ablation) | BokehMe neural renderer · BLB Level 3 | PSNR: 39.19 -> 39.21 | Improved |
Later papers that compared with this method
2 later accepted papers report BokehMe as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Dr. Bokeh: DiffeRentiable Occlusion-aware Bokeh Rendering (CVPR 2024 · better on 5 of 5 shared results)
- DC2: Dual-Camera Defocus Control by Learning To Refocus (CVPR 2023 · better on 1 of 1 shared result)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- BokehMe++: Harmonious Fusion of Classical and Neural Rendering for Versatile Bokeh Creation (TPAMI 2024 · similarity 0.95)
- BokehFlow: Depth-Free Controllable Bokeh Rendering via Flow Matching (AAAI 2026 · similarity 0.90)
- BokehCrafter: Taming Video Diffusion Models for Controllable Bokeh Rendering (AAAI 2026 · similarity 0.90)
- Dr. Bokeh: DiffeRentiable Occlusion-aware Bokeh Rendering (CVPR 2024 · similarity 0.89)
- AnyBokeh: Physics-Guided Any-to-Any Bokeh Editing with Optical Fingerprint Transfer (NeurIPS 2026 · similarity 0.85)
- HyRF: Hybrid Radiance Fields for Memory-efficient and High-quality Novel View Synthesis (NeurIPS 2025 · similarity 0.85)
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.
