Calibration models
2 calibration models. Each section says what the model is, whether it may be used commercially and how fast it is; licences and checks are under Details.
Input: one RGB image. Output: the camera intrinsics [3, 3] and, where the model estimates them, the gravity direction in the camera frame, roll, pitch and vertical field of view.
Measured on RTX 5060 Ti, torch 2.14.1+cu130; fastforks bench MODEL --size HxW measures yours. The summary uses one input, 1080x1920: kernels time with its speed-up over the faster original run, and kernels peak reserved memory with its ratio to the smaller original one. Where measured (one variant per family), a variant's table also lists portable under torch.compile and the fp16 TensorRT engine of portable (export; memory: the engine's device memory + torch's peak). – = not measured; no kernels = the family has no kernels backend (reason under its table). Reserved memory is the allocator's cache: kernels (CUDA graphs, cuDNN workspaces) can reserve a few percent to some tens of percent more than portable at the same allocated peak.
Other devices: Apple M1 Pro 16 GB, NVIDIA L4, NVIDIA L40S, NVIDIA A10, NVIDIA A100-SXM4-40GB, NVIDIA RTX PRO 6000; across devices.
| model | commercial | kernels time (ms), speed-up vs original | kernels peak memory (MiB, reserved), ratio to original |
|---|---|---|---|
anycalib-pinhole | yes | 21.5, 1.8x | 812, 0.6x |
geocalib-pinhole | yes | 81.8, 1.3x | 338, 0.61x |
All models
Modes, licences, training data and outputs per model
Commercial status per variant: yes if the code and the weights licence both allow commercial use (the training data is listed but does not change it; weights without a licence of their own are assumed to carry the code licence); no if either forbids it; unclear for custom licence terms (treat as no).
| model | modes | code licence | weights licence | training data | commercial | outputs |
|---|---|---|---|---|---|---|
anycalib-pinhole | portable, kernels | Apache-2.0 | Apache-2.0 | lvd-142m, openpano | yes | intrinsics, vfov |
geocalib-pinhole | portable, kernels | Apache-2.0 | CC-BY-4.0 | openpano | yes | intrinsics, gravity, roll, pitch, vfov |
Training datasets
| id | dataset | licence | commercial use |
|---|---|---|---|
lvd-142m | LVD-142M (DINOv2's curated web-image set, not released) | not stated | unclear |
openpano | OpenPano (perspective crops of public 360-degree panoramas: PolyHaven, HDRMaps, AmbientCG, BlenderKit) | not stated | unclear |