Each ranking includes only the best model for one method.
- UnrealStereo4K (3840×2160): AbsRel<=0.04
- MVS-Synth (1920×1080): AbsRel<=0.06
- HRSD (1920×1080): AbsRel<=0.08
- Middlebury2021 (1920×1080): SqRel<=0.5
- NYU-Depth V2 (640×480): OPW<=0.31
- NYU-Depth V2 (640×480): AbsRel<=0.058
- (to do)
- 👑 4DLFVD with up to 10×10 real light field views✔️: LPIPS😍 (no data)
This will be the King of all rankings. We look forward to ambitious researchers. - 4DLFVD with up to 10×10 real light field views✔️: PSNR😞 (no data)
- Hybrid with 7×7 synthetic light field views✖️: LPIPS😍 (no data)
- Hybrid with 7×7 synthetic light field views✖️: PSNR😞>=32dB
- Appendix 1: Rules for qualifying models for the rankings (to do)
- Appendix 2: Metrics selection for the rankings (to do)
- Appendix 3: List of all research papers from the above rankings
| RK | Model | AbsRel ↓ {Input fr.} |
Training dataset |
Official repository |
Practical model |
VapourSynth |
|---|---|---|---|---|---|---|
| 1 | ZoeDepth +PFR=128 ENH:  |
0.0388 {1} |
ENH: UnrealStereo4K |
 ENH:  |
- | - |
| RK | Model | AbsRel ↓ {Input fr.} |
Training dataset |
Official repository |
Practical model |
VapourSynth |
|---|---|---|---|---|---|---|
| 1 | ZoeDepth +PFR=128 ENH: |
0.0589 {1} |
ENH: MVS-Synth |
ENH: |
- | - |
| RK | Model | AbsRel ↓ {Input fr.} |
Training dataset |
Official repository |
Practical model |
VapourSynth |
|---|---|---|---|---|---|---|
| 1 | DPT-B + R + AL ENH:  |
0.074 {1} |
ENH: HRSD |
 ENH: - |
- | - |
| RK | Model | SqRel ↓ {Input fr.} |
Training dataset |
Official repository |
Practical model |
VapourSynth |
|---|---|---|---|---|---|---|
| 1 | LeReS-GBDMF ENH:  |
0.444 {1} |
ENH: HR-WSI |
 ENH:  |
- | - |
| RK | Model | OPW ↓ {Input fr.} |
Training dataset |
Official repository |
Practical model |
VapourSynth |
|---|---|---|---|---|---|---|
| 1 | FutureDepth Backbone: Swin-L |
0.303 {4} |
NYU-Depth V2 | - | - | - |
| RK | Model | AbsRel ↓ {Input fr.} |
Training dataset |
Official repository |
Practical model |
VapourSynth |
|---|---|---|---|---|---|---|
| 1 | Metric3D v2 CSTM_label ENH: Backbone: DINOv2 with registers (ViT-L/14) |
0.042 {1} |
DDAD & Lyft & Driving Stereo & DIML & Arogoverse2 & Cityscapes & DSEC & Mapillary PSD & Pandaset & UASOL & Virtual KITTI & Waymo & Matterport3d & Taskonomy & Replica & ScanNet & HM3d & Hypersim |  |
- | - |
| 2 | Depth Anything Large Backbone: DINOv2 (ViT-L/14) |
0.043 {1} |
Pretraining: BlendedMVS & DIML & HR-WSI & IRS & MegaDepth & TartanAir Training: BDD100K & Google Landmarks & ImageNet-21K & LSUN & Objects365 & Open Images V7 & Places365 & SA-1B |
 |
- | - |
| 3 | MiDaS v3.1 BEiTL-512 ENH: Backbone: BEiT512-L (ViT-L/16) |
0.048 {1} |
Pretraining: ReDWeb & HR-WSI & BlendedMVS & NYU-Depth V2 & KITTI Training: ReDWeb & DIML & 3D Movies & MegaDepth & WSVD & TartanAir & HR-WSI & ApolloScape & BlendedMVS & IRS & NYU-Depth V2 & KITTI |
 |
- |   |
| 4 | Marigold Backbone: text-to-image LDM (Stable Diffusion v2) |
0.055 {1} |
Hypersim & Virtual KITTI |  |
- | - |
| 5 | GenPercept Backbone: VAE & U-Net (Stable Diffusion v2.1) |
0.056 {1} |
Hypersim & Virtual KITTI |  |
- | - |
| 6 | NeWCRFs + LightedDepth ENH:  |
0.057 {2} |
ENH: NYU-Depth V2 |
 ENH:  |
- | - |
| 7 | UniDepth-V Backbone: DINOv2 (ViT-L/14) |
0.0578 {1} |
A2D2 & Argoverse2 & BDD100k & CityScapes & DrivingStereo & Mapillary PSD & ScanNet & Taskonomy & Waymo |  |
- | - |
| RK | Model | PSNR ↑ {Input fr.} |
Training dataset |
Official repository |
Practical model |
VapourSynth |
|---|---|---|---|---|---|---|
| 1 | LFVRT MDE: DPT Backbone: ViT |
32.66 {3+1D} |
GoPro & TAMULF |  MDE: |
- | - |
📝 Note: The above ranking includes only one model, as the other methods are image-based and don't have any temporal information making them unsuitable for light field video reconstruction from monocular video.
| Method | Paper | Venue |
|---|---|---|
| Depth Anything | Depth Anything: Unleashing the Power of Large-Scale Unlabeled Data | |
| DPT | Vision Transformers for Dense Prediction | |
| FutureDepth | FutureDepth: Learning to Predict the Future Improves Video Depth Estimation | |
| GBDMF | Multi-Resolution Monocular Depth Map Fusion by Self-Supervised Gradient-Based Composition | |
| GenPercept | Diffusion Models Trained with Large Data Are Transferable Visual Models | |
| LeReS | Learning to Recover 3D Scene Shape from a Single Image | |
| LightedDepth | LightedDepth: Video Depth Estimation in light of Limited Inference View Angles | |
| LFVRT | Synthesizing Light Field Video from Monocular Video | |
| Marigold | Repurposing Diffusion-Based Image Generators for Monocular Depth Estimation | |
| Metric3D | Metric3D: Towards Zero-shot Metric 3D Prediction from A Single Image | |
| Metric3D v2 | Metric3D v2: A Versatile Monocular Geometric Foundation Model for Zero-shot Metric Depth and Surface Normal Estimation | |
| MiDaS | Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-Shot Cross-Dataset Transfer | |
| MiDaS v3.1 | MiDaS v3.1 – A Model Zoo for Robust Monocular Relative Depth Estimation | |
| NeWCRFs | Neural Window Fully-connected CRFs for Monocular Depth Estimation | |
| PatchFusion | PatchFusion: An End-to-End Tile-Based Framework for High-Resolution Monocular Metric Depth Estimation | |
| R + AL | High-Resolution Synthetic RGB-D Datasets for Monocular Depth Estimation | |
| UniDepth | UniDepth: Universal Monocular Metric Depth Estimation | |
| ZoeDepth | ZoeDepth: Zero-shot Transfer by Combining Relative and Metric Depth | |