Research map / 3D vision / Depth and stereo
Neural light field and illumination imaging
Recovers scene properties with learned light fields, structured light and material estimation.
107 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2016–2026) · an approach in Depth and stereo, keyword cluster “light · imaging · illumination”
State History Tried Preprints Who Export
Where it stands
Not enough benchmark data Fewer than four papers report the same benchmark, so there is no record to follow.
26 papers in 2023–24 and 19 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 8.5% to 4.6%.
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2016 · Convolutional Networks for Shape From Light Field (CVPR 2016 · 118 citations)
- 2017 · Deep Outdoor Illumination Estimation (CVPR 2017 · 234 citations)
- 2018 · EPINET: A Fully-Convolutional Neural Network Using Epipolar Geometry for Depth From Light Field Images (CVPR 2018 · 317 citations)
- 2019 · DeepView: View Synthesis With Learned Gradient Descent (CVPR 2019 · 434 citations)
- 2020 · Single-Image HDR Reconstruction by Learning to Reverse the Camera Pipeline (CVPR 2020 · 285 citations)
- 2021 · Deep Polarization Imaging for 3D Shape and SVBRDF Acquisition (CVPR 2021 · 85 citations)
- 2022 · Light Field Neural Rendering (CVPR 2022 · 112 citations)
- 2023 · ShadowNeuS: Neural SDF Reconstruction by Shadow Ray Supervision (CVPR 2023 · 25 citations)
- 2024 · Progressive Divide-and-Conquer via Subsampling Decomposition for Accelerated MRI (CVPR 2024 · 16 citations)
- 2025 · Neural LightRig: Unlocking Accurate Object Normal and Material Estimation with Multi-Light Diffusion (CVPR 2025 · 7 citations)
- 2026 · Phys-Liquid: A Physics-Informed Dataset for Estimating 3D Geometry and Volume of Transparent Deformable Liquids (AAAI 2026 · 2 citations)
Earliest papers here
- Convolutional Networks for Shape From Light Field (CVPR 2016 · 118 citations)
- Deep Reflectance Maps (CVPR 2016 · 96 citations)
- What Sparse Light Field Coding Reveals About Scene Structure (CVPR 2016 · 92 citations)
What has been tried
93 papers here report 829 measured changes (a component added, removed or swapped, compared with a baseline): helped 77%, no clear effect 4%, hurt 17%, mixed 2%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| add adaptive degradation prediction | Stanford2D FSE · PSNR | 36.46 -> 36.77 | Helped | Progressive Divide-and-Conquer via Subsampling Decomposition for Accelerated MRI (CVPR 2024 · 16 citations) |
| Add polarization cues | Bunny · normal MAE | 3.295° -> 1.727° | Helped | NeISF: Neural Incident Stokes Field for Geometry and Material Estimation (CVPR 2024 · 15 citations) |
| add projector loss to the camera-loss-only setup | Statue reconstruction, three a la carte patterns · reconstruction fidelity | Helped | TurboSL: Dense Accurate and Fast 3D by Neural Inverse Structured Light (CVPR 2024 · 9 citations) | |
| add core factor V | Room and Scuol · PSNR | 29.74 -> 30.20; 27.59 -> 27.84 | Helped | NeLF-Pro: Neural Light Field Probes for Multi-Scale Novel View Synthesis (CVPR 2024 · 8 citations) |
| add reference attention to concatenation | Multi-light diffusion · PSNR | 19.32 -> 20.01 | Helped | Neural LightRig: Unlocking Accurate Object Normal and Material Estimation with Multi-Light Diffusion (CVPR 2025 · 7 citations) |
| Use adaptive template generation with k=4 rather than a straight initial template | HaN-Seg · Dice | 78.87 -> 79.58 | Helped | DeformCL: Learning Deformable Centerline Representation for Vessel Extraction in 3D Medical Image (CVPR 2025 · 6 citations) |
| add the view2view framework | Synthetic dataset (Average) · PSNR | 30.81 -> 39.05 | Helped | V2V3D: View-to-View Denoised 3D Reconstruction for Light Field Microscopy (CVPR 2025 · 5 citations) |
| Add joint volume and decomposed surface rendering | SK3D Pot · Chamfer distance | 2.01 -> 1.88 | Helped | Factored-NeuS: Reconstructing Surfaces, Illumination, and Materials of Possibly Glossy Objects (CVPR 2025 · 4 citations) |
| add emitter mask Me | FIPT synthetic scenes · kd PSNR | 16.50 -> 22.33 | Helped | IRIS: Inverse Rendering of Indoor Scenes from Low Dynamic Range Images (CVPR 2025 · 4 citations) |
| use octree depth 7 instead of depth 6 | FutureHouse · PSNR | 15.34 -> 17.14 | Helped | LightOctree: Lightweight 3D Spatially-Coherent Indoor Lighting Estimation (CVPR 2024 · 3 citations) |
| Jointly optimize the structured-light patterns with the reconstruction network instead of using fixed random binary or Hadamard patterns | Synthetic reconstruction with 1 camera · Reconstruction quality | Helped | Real-time Acquisition and Reconstruction of Dynamic Volumes with Neural Structured Illumination (CVPR 2024 · 2 citations) | |
| add ACES tone mapping | Synthetic scenes (average) · PSNR (albedo) | 21.24 -> 25.09 | Helped | RobIR: Robust Inverse Rendering for High-Illumination Scenes (NeurIPS 2024 · 2 citations) |
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 45 papers; from OpenAlex and the papers' affiliations).
- Australian National University · 3 papers
- Nanyang Technological University · 3 papers
- Zhejiang University · 3 papers
- Beihang University · 2 papers
- University of Science and Technology of China · 2 papers
- Tencent · 2 papers
- Tsinghua University · 2 papers
Use it
Download BibTeX of 40 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Non-line-of-sight imaging (93 papers)
- Spike and event camera reconstruction (50 papers)
- Snapshot hyperspectral imaging (38 papers)
Elsewhere in the topic
- Feed-forward 3D reconstruction (133 papers)
- Self-supervised monocular depth (116 papers)
- Scene coordinate regression for localization (112 papers)
- Learned optical and scene flow (108 papers)
- Cost-volume multi-view stereo (107 papers)
- Camera model and calibration for SfM (99 papers)
- Dense depth prediction and completion (85 papers)
- Foundation models for monocular geometry (81 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).