Research map / 3D vision / Depth and stereo
AnyCalib: On-Manifold Learning for Model-Agnostic Single-View Camera Calibration
Javier Tirado-Garín, Javier Civera
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 FoV fields as target instead of rays with cosine loss (ablation) | AnyCalib_pinhole · Averaged perspective benchmarks | RE: 26.05 -> 23.81 | Improved |
| use FoV fields instead of DSINE-style angular ray loss (ablation) | AnyCalib_pinhole · Averaged perspective benchmarks | RE: 24.95 -> 23.81 | Improved |
| use light DPT decoder without transposed convolutions (ablation) | AnyCalib_pinhole · Averaged perspective benchmarks | MACs: 243.2G -> 187.8G | Improved |
| light DPT decoder instead of original (ablation) | AnyCalib_pinhole · Averaged perspective benchmarks | RE: 24.99 -> 23.81 | Improved |
| extend OpenPano with 4055 unconstrained panoramas (ablation) | AnyCalib_pinhole · Averaged perspective benchmarks | RE: 26.62 -> 23.81 | Improved |
| fit intrinsics with RANSAC minimal samples instead of non-minimal closed form (ablation) | AnyCalib on Kannala-Brandt · ScanNet++ | RE: 20.61 -> 1019 | Hurt |
| train AnyCalib on general projections versus perspective-only for perspective images | AnyCalib_gen vs AnyCalib_pinhole · Stanford2D3D | vFoV error (deg): 2.55 -> 2.95 | Mixed |
| swap camera model at runtime (radial, KB, UCM) without retraining | AnyCalib · Distorted images | AE (deg) | No clear effect |
| compare AnyCalib with 3D foundation model MoGe on perspective images | AnyCalib_pinhole · TartanAir | vFoV error (deg): 1.61 -> 3.62 | Hurt |
| apply AnyCalib to edited (stretched/cropped) images versus DiffCalib | AnyCalib · TartanAir (edited) | e_f (relative focal error): 0.62 -> 0.19 | Improved |
Later papers that compared with this method
1 later accepted paper reports AnyCalib as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- InFlux++: Real and Synthetic Data for Estimating Dynamic Camera Intrinsics (ECCV 2026 · better on 5 of 14 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 4 on most shared results.
- GeoCalib — GeoCalib: Learning Single-image Calibration with Geometric Optimization (ECCV 2024) · better on 5 of 5 shared results
- MoGe — MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervision (CVPR 2025) · better on 1 of 4 shared results
- DiffCalib — DiffCalib: Reformulating Monocular Camera Calibration as Diffusion-Based Dense Incident Map Generation (AAAI 2025) · better on 2 of 2 shared results
- DUSt3R — DUSt3R: Geometric 3D Vision Made Easy (CVPR 2024) · better on 1 of 1 shared result
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
- AnyCam: Learning to Recover Camera Poses and Intrinsics from Casual Videos (CVPR 2025 · similarity 0.87)
- AnyMap: Learning a General Camera Model for Structure-from-Motion with Unknown Distortion in Dynamic Scenes (CVPR 2025 · similarity 0.87)
- DiffCalib: Reformulating Monocular Camera Calibration as Diffusion-Based Dense Incident Map Generation (AAAI 2025 · similarity 0.86)
- Minimal Perspective Autocalibration (CVPR 2024 · similarity 0.85)
- LoRA3D: Low-Rank Self-Calibration of 3D Geometric Foundation models (ICLR 2025 · similarity 0.85)
- InFlux: A Benchmark for Self-Calibration of Dynamic Intrinsics of Video Cameras (NeurIPS 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.
