Research map / 3D vision / Depth and stereo
MDP-Omni: Parameter-free Multimodal Depth Prior-based Sampling for Omnidirectional Stereo Matching
Eunjin Son, HyungGi Jo, Wookyong Kwon, Sang Jun Lee
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 azimuth-based multi-view volume fusion (AMVF, k=3) (ablation) | MDP-Omni · OmniThings | MAE: 1.27 -> 1.20 | Improved |
| add azimuth-based multi-view volume fusion (AMVF, k=3) instead of variance fusion (ablation) | MDP-Omni · OmniThings | MAE: 1.25 -> 1.20 | Improved |
| use steepness k=3 instead of k=1 (ablation) | AMVF · OmniThings | MAE: 1.23 -> 1.20 | Improved |
| use k=3 instead of equal weights (k=inf) (ablation) | AMVF · OmniThings | MAE: 1.21 -> 1.20 | Improved |
| add 3D positional embedding (ablation) | MDP-Omni with DPS · OmniThings | MAE: 1.27 -> 1.20 | Improved |
| add 3D positional embedding (ablation) | Unimodal sampling · OmniThings | MAE: 1.69 -> 1.64 | Improved |
| replace unimodal sampling with DPS (ablation) | MDP-Omni · OmniThings | MAE: 1.64 -> 1.20 | Improved |
| replace unimodal sampling with DPS (ablation) | MDP-Omni · OmniThings | >3 error: 10.47 -> 7.09 | Improved |
| replace top-K sampling with DPS (ablation) | MDP-Omni · OmniThings | MAE: 1.28 -> 1.20 | Improved |
| replace top-K sampling with DPS (ablation) | MDP-Omni · OmniThings | RMS: 4.01 -> 3.79 | Improved |
| full MDP-Omni vs OmniMVS+32 (pretrained) | MDP-Omni · OmniThings | MAE: 1.37 -> 1.20 | Improved |
| full MDP-Omni vs RomniStereo32 (pretrained) | MDP-Omni · OmniHouse | MAE: 0.80 -> 1.16 | Hurt |
Later papers that compared with this method
2 later accepted papers report MDP-Omni as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- OmniDS: Dual-Stream Context Fusion for Omnidirectional Depth from Fisheye Cameras (ECCV 2026 · better on 4 of 4 shared results)
- PDF-Omni: Poincaré Dual Disk Distortion Field-based Recurrent Update for Omnidirectional Stereo Matching (ECCV 2026 · better on 3 of 3 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- OmniMVS+32 — OmniMVS: End-to-End Learning for Omnidirectional Stereo Matching (ICCV 2019) · better on 4 of 4 shared results
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
- PDF-Omni: Poincaré Dual Disk Distortion Field-based Recurrent Update for Omnidirectional Stereo Matching (ECCV 2026 · similarity 0.91)
- End-to-End Learning for Omnidirectional Stereo Matching With Uncertainty Prior (TPAMI 2020 · similarity 0.90)
- OmniDS: Dual-Stream Context Fusion for Omnidirectional Depth from Fisheye Cameras (ECCV 2026 · similarity 0.89)
- OmniStereo: Real-time Omnidireactional Depth Estimation with Multiview Fisheye Cameras (CVPR 2025 · similarity 0.87)
- MonSter: Marry Monodepth to Stereo Unleashes Power (CVPR 2025 · similarity 0.86)
- HELVIPAD: A Real-World Dataset for Omnidirectional Stereo Depth Estimation (CVPR 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.
