Research map / 3D vision / Face and shape
Geometry Fidelity for Spherical Images
Anders Christensen, Nooshin Mojab, Khushman Patel, Karan Ahuja, Zeynep Akata, Ole Winther, Mar Gonzalez Franco, Andrea Colaco
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 |
|---|---|---|---|
| use cubemap-based OmniFID instead of FID on equirectangular images to detect reduced field of view | FID metric · 360-Indoor, FOV 180 to 140 degrees | sensitivity to geometry: FID only 10 under FOV reduction | Improved |
| apply OmniFID instead of FID across fine-tuning checkpoints | Imagen fine-tuned on 360-Indoor · generated panoramas | OmniFID vs FID trend: OmniFID 63.39 -> 60.38 -> 55.07; FID 33.96 -> 35.42 -> 34.95 | Improved |
| extend FID to cubemap with OmniFID under unrelated noise (salt and pepper, Gaussian, blur) | OmniFID · 360-Indoor noise tests | noise sensitivity | No clear effect |
| use relative edge change across seam instead of plain average edge magnitude in Discontinuity Score (ablation) | Discontinuity Score · seam scoring | detection of abrupt semantic stops | Improved |
| corrupt Laval images by cropping 0.25% of width on each border | Discontinuity Score · Laval corrupted | DS | Improved |
| group cubemap views into frontal, up and down before averaging features (ablation) | OmniFID · 360-Indoor | feature mean distance: 0.34 within frontal vs 24.27 and 33.13 to up/down | Improved |
Closest papers by meaning
- SphereDiffusion: Spherical Geometry-Aware Distortion Resilient Diffusion Model (AAAI 2024 · similarity 0.86)
- Spherical Pseudo-Cylindrical Representation for Omnidirectional Image Super-resolution (AAAI 2024 · similarity 0.85)
- SphereUFormer: A U-Shaped Transformer for Spherical 360 Perception (CVPR 2025 · similarity 0.85)
- ODGS: 3D Scene Reconstruction from Omnidirectional Images with 3D Gaussian Splattings (NeurIPS 2024 · similarity 0.84)
- UniK3D: Universal Camera Monocular 3D Estimation (CVPR 2025 · similarity 0.84)
- OmniSDF: Scene Reconstruction using Omnidirectional Signed Distance Functions and Adaptive Binoctrees (CVPR 2024 · 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.
