Research map / 3D vision / Depth and stereo
PDF-Omni: Poincaré Dual Disk Distortion Field-based Recurrent Update for Omnidirectional Stereo Matching
Yunseok Yang, Eunjin Son, Sang 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 DAA+DiGRU (distortion-informed GRU) (ablation) | baseline RAFT-style OSM · Sunny (trained on OmniThings only) | >1 error (%): 9.41 -> 8.19 | Improved |
| add LEC loss to DAA+DiGRU (ablation) | DAA+DiGRU · OmniThings (pretrained) | >3 error (%): 6.85 -> 6.73 | Improved |
| add LEC loss when GE and DAA+DiGRU are combined (ablation) | DAA+DiGRU + GE · Sunny (trained on OmniThings only) | RMS: 3.52 -> 2.42 | Improved |
| replace ConvGRU with Selective-Stereo SRU (ablation) | baseline · OmniThings (pretrained) | >1 error (%): 16.52 -> 17.03 | Hurt |
| use Euclidean instead of Poincare distortion field (ablation) | DiGRU+DAA · Sunny (trained on OmniThings only) | >1 error (%): 8.61 -> 14.44 | Hurt |
| use Euclidean instead of Poincare distortion field (ablation) | DiGRU+DAA · OmniHouse (trained on OmniThings only) | RMS: 2.23 -> 1.98 (OmniHouse RMS lower; Sunny worse) | Mixed |
| invert the distortion map (1-D) (ablation) | DiGRU+DAA · OmniThings (pretrained) | >1 error (%): 14.86 -> 15.90 | Hurt |
| drop in DAA+DiGRU replacing GRU (ablation) | RomniStereo32 · OmniThings (pretrained) | >1 error (%): 20.42 -> 17.27 | Improved |
| drop in Selective-Stereo SRU replacing GRU (ablation) | RomniStereo32 · OmniThings (pretrained) | >1 error (%): 20.42 -> 18.80 | Improved |
| PDF-Omni vs MDP-Omni, fine-tuned | PDF-Omni-ft · Sunny seam region | >1 error (%): 5.879 -> 4.314 | Improved |
| PDF-Omni vs prior best, pretrained on OmniThings | PDF-Omni · OmniHouse (trained on OmniThings only) | MAE: 0.80 (RomniStereo32) vs 0.86 | No clear effect |
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.
- MDP-Omni — MDP-Omni: Parameter-free Multimodal Depth Prior-based Sampling for Omnidirectional Stereo Matching (ICCV 2025) · better on 3 of 3 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
- MDP-Omni: Parameter-free Multimodal Depth Prior-based Sampling for Omnidirectional Stereo Matching (ICCV 2025 · similarity 0.91)
- OmniDS: Dual-Stream Context Fusion for Omnidirectional Depth from Fisheye Cameras (ECCV 2026 · similarity 0.89)
- End-to-End Learning for Omnidirectional Stereo Matching With Uncertainty Prior (TPAMI 2020 · similarity 0.88)
- OmniStereo: Real-time Omnidireactional Depth Estimation with Multiview Fisheye Cameras (CVPR 2025 · similarity 0.86)
- Rectification-specific Supervision and Constrained Estimator for Online Stereo Rectification (CVPR 2025 · similarity 0.86)
- MonSter: Marry Monodepth to Stereo Unleashes Power (CVPR 2025 · similarity 0.84)
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.
