AnyCalib (fastforks)
AnyCalib (Tirado-GarΓn and Civera, ICCV 2025): single-image camera calibration. A DINOv2 ViT-L/14 with a light DPT decoder predicts one ray per pixel (tangent coordinates on the unit sphere, convex upsampling); a linear fit and Gauss-Newton refinement give the intrinsics of the chosen camera model. Gives the pinhole intrinsics (fx, fy, cx, cy) and the vertical field of view; no gravity.
This repo is the model card only: it holds no weights. fastforks downloads the authors' weights from the links below and checks them against the SHA-256 in the model card. Code: https://github.com/kleinicke/fastforks. Kernels for the kernels backend: https://huggingface.co/fastforks/kernels. Original code: https://github.com/javrtg/AnyCalib at 027a8497d893.
Task: calibration. Tier: experimental. Outputs: intrinsics, vfov. Backends: portable, kernels.
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.
Variants
Weights are the authors' files (not hosted here). SHA-256 shortened to 12 hex digits; the full hashes are in card.toml.
| variant | weights (author link) | SHA-256 | weights licence | commercial use |
|---|---|---|---|---|
anycalib-pinhole@1 |
anycalib_pinhole.pt | e73b174563bc |
Apache-2.0 | yes |
Licence
Original-code strings, recorded as they are (research-use package; check them before any other use). Code: Apache-2.0. Licence files: LICENSE.
Training data: LVD-142M (DINOv2's curated web-image set, not released), OpenPano (perspective crops of public 360-degree panoramas: PolyHaven, HDRMaps, AmbientCG, BlenderKit) (anycalib_pinhole: DINOv2 ViT-L/14 (LVD-142M) fine-tuned on the perspective OpenPano split (OP_p; README, paper Sec. 5). Only the weights licence counts for the commercial flag.).
Licence evidence: https://github.com/javrtg/AnyCalib/blob/027a8497d893f4b2596f23d6324c05e4b81064ed/LICENSE ; https://github.com/javrtg/AnyCalib/blob/027a8497d893f4b2596f23d6324c05e4b81064ed/README.md
Input size
any H x W >= 32, padded to a multiple of 32
Tested up to (portable, fp16, RTX 5060 Ti 16 GB): anycalib-pinhole 3000x4000 (1.0 GiB peak allocated, 1.1 reserved)
Speed and memory
Calibration, synthetic, 1080x1920
| model | time (ms): original | portable | kernels | kernels speed-up: vs original | vs portable | peak memory (MiB, allocated / reserved): original | portable | kernels |
|---|---|---|---|---|---|---|---|---|
anycalib-pinhole |
38.4 | 29.4 | 21.3 | 1.8x | 1.4x | 1343 / 1352 | 705 / 742 | 741 / 816 |
Startup and first call
| model | load (s): original | portable | kernels | first call (s): original | portable | kernels |
|---|---|---|---|---|---|---|
anycalib-pinhole |
4.09 | 0.60 | 1.59 | 0.58 | 0.45 | 0.33 |
torch.compile and TensorRT
| model | input | time: portable | compile | TensorRT fp16 | kernels | preparation: compile (cold first call) | ONNX + engine build | peak MiB (alloc / reserved): portable | compile | kernels | TensorRT (engine + torch) |
|---|---|---|---|---|---|---|---|---|---|---|---|
anycalib-pinhole |
1080x1920 | 29.4 ms | 25.5 ms | planned for 0.2.1 | 21.3 ms | 16.5 s | β | 705 / 742 | 719 / 756 | 741 / 816 | β |
Fixed-size cells: time (ms) / peak allocated / peak reserved memory (MiB).
Hardware, protocol and caveats: benchmarks (RTX 5060 Ti 16 GB, sm_120). Original = the original code named above at the precision of portable (fp16, or (fp32) where fp16 is unsupported or not finite); original fp32 = its default fp32 run. Peak memory is the torch allocator's.
Accuracy
On public datasets
Each image at its own size (1440 x 1920 portrait); roll, pitch and vertical field of view of the fitted camera against the ground truth; AUC of the error up to 5 degrees (errors below 1 degree count as 1, siclib's compute_auc(min_error=1)), the GeoCalib paper's protocol. Original = the original code named under Details, same checkpoint and protocol.
| variant | dataset | original fp32 | original fp16 | portable | kernels | note |
|---|---|---|---|---|---|---|
anycalib-pinhole |
LaMAR2k, all 2000 images, roll / pitch / vertical FoV error AUC@5 degrees % | β / β / 51.6 | β / β / 51.6 | β / β / 51.5 | ||
anycalib-pinhole |
LaMAR2k, every 10th image (200), roll / pitch / vertical FoV error AUC@5 degrees % | β / β / 48.5 | β / β / 48.5 | β / β / 48.4 | β / β / 48.4 |
Measured on NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 6637f957, 2026-11-13.
Parity with the original code
Every variant is checked against reference outputs of the original code at its pinned commit. CPU, float32: max |diff| <= 0.00019 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.094 and p99 <= 0.51 (absolute error of the output intrinsics).
Usage
from fastforks import calibration
model = calibration.load("anycalib-pinhole") # mode="auto": the preferred backend from recorded measurements
result = model("photo.jpg")
result.intrinsics # [3, 3] float32 K in input pixels
# result.roll, result.pitch, result.vfov (radians), result.gravity [3]: where the model estimates them
result.backend # what ran, e.g. Backend(mode='portable', device='cuda:0', dtype='float16')
Install with pip install fastforks (see https://github.com/kleinicke/fastforks). Other variants: anycalib-pinhole.
Kernels backend
calibration.load("anycalib-pinhole", mode="kernels") runs our CUDA kernels (fastbb, dpt_fused, cudnn_hacc). It needs a prebuilt kernel for your GPU architecture (downloaded from https://huggingface.co/fastforks/kernels on first use) and a recent NVIDIA driver; if there is none it raises BackendUnavailableError with the reason and never falls back silently. mode="auto" picks kernels when it can run and logs the choice.