LoFTR
LoFTR (Zhejiang University, CVPR 2021): detector-free matching of two photos of one scene with transformers; semi-dense matches as keypoint pairs (the form COLMAP takes). Outdoor (MegaDepth) and indoor (ScanNet) weights. Apache-2.0 code and weights.
experimental Fork of zju3dv/LoFTR@df7ca80 · backends: portable, kernels
- Paper LoFTR: Detector-Free Local Feature Matching with Transformers CVPR 2021 · first released 1 Apr 2021
Paper 2021 Added in 0.2.0 Evaluated with 0.2.0
Use it
pip install fastforks Lean install with only what this model needs: pip install fastforks-core
from fastforks import matching
model = matching.load("loftr-outdoor") # kernels on a supported GPU, else portable PyTorch
result = model(image_a, image_b)
result.keypoints1 # also: keypoints2, pairs, scores
result.backend # what ran: mode, device, dtype, kernels
points1, points2 = result.matches(5000) # matched pixels, sampled where the model is certain
pose = matching.estimate_pose(points1, points2, K1, K2) # relative pose on the GPU (RANSAC); K = intrinsics [3, 3]
pose.R, pose.t, pose.inliers # rotation, unit translation, inlier mask fastforks bench loftr-outdoor Times the model on this machine. load(…, mode="portable") or mode="kernels" picks a backend; usage guide.
Speed and memory
Every backend on one GPU (RTX 5060 Ti 16 GB) and the same input, the authors’ code included. Kernels runs fp16. Portable runs fp16 on CUDA.
Input: synthetic, 768x1024 (the task’s standard size)
| Backend | dtype | Time per call (ms) | vs original fp16 | Peak memory allocated / reserved | Load (s) | First call (s) |
|---|---|---|---|---|---|---|
| Original fp32 | fp32 | not measured | – | – | – | – |
| Original (own default)upstream LoFTR default_cfg, temp_bug_fix False, long side 840 | fp32 | 241 | – | 1,739 MiB/1,870 MiB | 0.3 | 0.24 |
| Portable fp32 | fp32 | not measured | – | – | – | – |
| Portable | fp16 | 110 | 2.2× | 796 MiB/904 MiB | 0.2 | 0.58 |
| Kernels | fp16 | 95.9 | 2.5× | 760 MiB/886 MiB | 0.7 | 0.37 |
| Portable + torch.compile | 176 | 1.4× | 1,149 MiB/1,232 MiB | 33 (build) | – | |
| TensorRTplanned for 0.2.1 | fails | – | – | – | – |
Measured on RTX 5060 Ti 16 GB, torch 2.14.1+cu130, 2026-11-13 to 2026-11-16. First call, 3 warm-up, peak reset, median of 3 rounds x median of 5. Ratios divide two
times of this table (same GPU, same input). Original fp16 is the authors’ own precision option, else torch.autocast. torch.compile and TensorRT: build time instead of load time. TensorRT memory (engine, not comparable) is the GPU memory the engine took, read from the driver because TensorRT allocates outside PyTorch, so it can't be compared with the PyTorch peaks of the other rows.
Other devices
| Device | Backend | dtype | ms / call | Peak memory | Output vs reference | Input |
|---|---|---|---|---|---|---|
| Apple M1 Pro 16 GB MPS, torch 2.14.1, 2026-11-17 | Original | fp32 | 997 | 6,998 MiB process | – | 768x1024 |
| Portable | fp16 | 723 | 3,614 MiB process | matches | 768x1024 | |
| Portable | fp32 | 875 | 5,767 MiB process | matches | 768x1024 | |
| Kernels | fp16 | not implemented | – | 768x1024 | ||
| Kernels | fp32 | not implemented | – | 768x1024 | ||
CUDA memory is torch’s peak allocated / reserved; Apple has no such counter, so the Mac rows show the process footprint (weights, Python and torch included). “Output vs reference” compares the device’s output with the stored golden output under the card’s tolerance; – = not checked on that device. Kernels “not implemented” = no kernels for this model on that backend yet: it runs portable there.
Licence and weights
Commercial use allowed
Our reading of the licences, not legal advice.
Weights
loftr_outdoor.ckpt 46.3 MB· hosted on Hugging Facefastforks downloads and checks these files on first use (fastforks download loftr-outdoor).
Accuracy
The test set through the authors’ code and through ours: same checkpoint, same protocol. The distance between the original’s fp32 and fp16 runs is what half precision alone costs; our runs should stay within it.
MegaDepth-1500
relative pose AUC@5 / @10 / @20 degrees
| Run | dtype | AUC@5° ↑ | AUC@10° ↑ | AUC@20° ↑ | Δ AUC@5° vs fp32 original | pairs | ms / pair |
|---|---|---|---|---|---|---|---|
| Original fp32 | fp32 | 50.1 | 67.5 | 80.3 | reference | 1500 | 152 |
| Original fp16 | not measured | ||||||
| Portable fp32 | fp32 | 50.8 | 67.8 | 80.2 | +0.6 | 1500 | 150 |
| Portable | fp16 | 50.6 | 67.6 | 80.3 | +0.5 | 1500 | 74.0 |
| Kernels | not measured | ||||||
Dense matchers: 5000 matches per pair sampled from the warp by certainty (upstream's balanced KDE sampling); sparse matchers: all their matches of one call (LightGlue, XFeat: LightGlue's MegaDepth-1500 setting; LoFTR, DeDoDe: their card defaults). OpenCV essential-matrix RANSAC (0.5 px threshold, normalised by the mean focal length); AUC of the larger of the rotation and translation-direction errors. Original = the authors’ code, same checkpoint and protocol. Measured on original fp32: RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-15; portable fp32, portable: RTX 5060 Ti, torch 2.14.1+cu130, fastforks 3a7e4880, 2026-11-15.
MegaDepth-1500
every 10th pair, relative pose AUC@5 / @10 / @20 degrees
| Run | dtype | AUC@5° ↑ | AUC@10° ↑ | AUC@20° ↑ | Δ AUC@5° vs fp32 original | pairs | ms / pair |
|---|---|---|---|---|---|---|---|
| Original fp32 | fp32 | 53.8 | 71.8 | 83.7 | reference | 150 | 155 |
| Original fp16 | not measured | ||||||
| Portable fp32 | fp32 | 52.1 | 71.0 | 83.3 | −1.7 | 150 | 155 |
| Portable | fp32 | 52.1 | 71.0 | 83.3 | −1.7 | 150 | 155 |
| Kernels | fp16 | 57.1 | 74.2 | 85.0 | +3.3 | 150 | 61.2 |
Dense matchers: 5000 matches per pair sampled from the warp by certainty (upstream's balanced KDE sampling); sparse matchers: all their matches of one call (LightGlue, XFeat: LightGlue's MegaDepth-1500 setting; LoFTR, DeDoDe: their card defaults). OpenCV essential-matrix RANSAC (0.5 px threshold, normalised by the mean focal length); AUC of the larger of the rotation and translation-direction errors. Original = the authors’ code, same checkpoint and protocol. Measured on original fp32: RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-13; portable fp32, portable: RTX 5060 Ti, torch 2.14.1+cu130, fastforks aee4b565, 2026-11-13; kernels: RTX 5060 Ti, torch 2.14.1+cu130, fastforks 68b2f345, 2026-11-16.
Relative pose on the GPU
A dense matcher is usually followed by RANSAC on its matches to get the camera motion. fastforks does that step on the
GPU too: result.matches(n) samples matched points where the model is certain, and matching.estimate_pose runs essential-matrix RANSAC on them, many image pairs per call, with no copy to
the CPU. Like the models it has a portable (plain PyTorch) and a kernels (CUDA) backend.
| Solver | Runs on | Threshold (px) | Pose AUC@5 / @10 / @20 | Pairs per second |
|---|---|---|---|---|
| fastforks estimate_pose, kernels | GPU, batches of 500 pairs | 1.0 | 69.9 / 81.5 / 89.3 | 18,446 |
| fastforks estimate_pose, portable | GPU, batches of 500 pairs | 1.0 | 69.6 / 81.2 / 89.1 | 418 |
| PoseLib 2.0.5 | CPU, 16 processes | 1.0 | 70.0 / 81.6 / 89.5 | 325 |
| OpenCV USAC (MAGSAC++) | CPU, 16 processes | 0.5 | 65.7 / 78.6 / 87.3 | 271 |
| OpenCV RANSAC | CPU, 16 processes | 0.5 | 62.9 / 76.9 / 86.4 | 323 |
MegaDepth-1500, RoMa v2 matches (5000 per pair), 5 samplings, the same matches for every solver, on one machine: RTX 5060 Ti and a Ryzen 7 5800X (16 threads). Each solver runs at the inlier threshold where it is most accurate. Pairs per second is wall-clock throughput over all 7500 poses: the CPU solvers run one pair per call in 16 single-threaded processes, the GPU solver takes 500 pairs per call. One pair alone takes about 1.7 ms with the kernels, 29 ms with PoseLib and 24 ms with OpenCV RANSAC on one core. The accuracy table above stays on OpenCV RANSAC, as in the papers, so it is comparable with published numbers; this table is the only place the GPU solver is used.
Closeness to the original
Beyond the test-set scores: the raw output of ours against the original’s fp32 output on the same input, pixel by pixel. CPU, float32: max |diff| <= 3e-05 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.0025 and p99 <= 0.02 (absolute error of the output keypoints1).
| Check | Tolerance | Measured |
|---|---|---|
| CPU fp32, max |diff| | ≤ 0.00003 | not recorded |
| cpu_min_recall | ≤ 0.99 | not recorded |
| cpu_min_match | ≤ 0.99 | not recorded |
| GPU default dtypes, mean |diff| | ≤ 0.0025 | 0.00211 |
| GPU default dtypes, p99 |diff| | ≤ 0.02 | 0.01445 |
| gpu_min_recall | ≤ 0.998 | not recorded |
| gpu_min_match | ≤ 0.998 | not recorded |
Recorded for loftr-outdoor on RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-15, with these kernels: cudnn_hacccutlass_convigev
Details
LoFTR (Sun et al., CVPR 2021): detector-free dense matching. A ResNet-FPN gives 1/8 and 1/2 resolution features; a linear-attention transformer matches the 1/8 cells of both images (dual-softmax, mutual nearest neighbours above a confidence threshold) and a second transformer refines each match to sub-pixel precision in a 5 x 5 window at 1/2 resolution. Returns the matched points of both images, their (identity) index pairs and the coarse confidence.
- Input / output
- Input: two RGB images of one scene. Output: for every pixel of the first image its position in the second (x, y in pixels) and a certainty that the pixel is visible in both.
What the numbers mean, and which models need the camera focal length
Sparse matchers (LightGlue) return the keypoints of both images and the index pairs of the matched ones instead. To reconstruct camera poses from an image folder with these matchers, see [structure from motion](../sfm.md).
- Input size
- any H x W >= 32, padded to a multiple of 32; tested up to 3000x4000 (1,943 MiB allocated)
- Options
long_side=840>= 0· changes the output
- Outputs
keypoints1,keypoints2,pairs,scores- Checkpoint
temp_bug_fix=false,border_rm=2- CUDA kernels
igevIGEV stereo: convolution epilogues, correlation lookups, GRU gatescutlass_convCUTLASS convolutions with the recurrent update (GRU) fused incudnn_haccconvolutions with fp16 accumulation (cuDNN)
- Apple (MPS)
- fp32, fp16
Limitations
- Sparse output: no dense warp. `result.matches()` gives the matched points for `estimate_pose`; every match is its own keypoint pair (`pairs` is `[[0, 0], [1, 1], ...]`).
- One image pair per call internally (batches run pair by pair); the match counts differ per pair.
- No exportable module: the match counts are data-dependent (`torch.compile` runs with graph breaks; ONNX / TensorRT export fails).
- Images are converted to grayscale and matched at `long_side` (default 840, upstream's MegaDepth test size); the coarse matching holds an (H W / 64)^2 score matrix, so memory grows with the 4th power of the side. The indoor weights were evaluated by upstream at 640 x 480 (ScanNet).
Full reference, generated from the repository
LoFTR (Zhejiang University, CVPR 2021): detector-free matching of two photos of one scene with transformers; semi-dense matches as keypoint pairs (the form COLMAP takes). Outdoor (MegaDepth) and indoor (ScanNet) weights. Apache-2.0 code and weights.
Commercial use: yes.
| variant | input | backend | time (ms) | peak memory (MiB, allocated / reserved) |
|---|---|---|---|---|
loftr-outdoor | synthetic (768x1024) | original portable kernels portable + torch.compile TensorRT fp16 | 241 110 95.9 176 planned for 0.2.1 | 1739 / 1870 796 / 904 760 / 886 1149 / 1232 – |
loftr-indoor | synthetic (768x1024) | original portable kernels portable + torch.compile TensorRT fp16 | 242 111 96.2 71.0 planned for 0.2.1 | 1751 / 1910 796 / 904 760 / 886 634 / 684 – |
Details
- Original code: https://github.com/zju3dv/LoFTR at
df7ca80f9173 - Code licence: Apache-2.0 (permissive); files:
LICENSE - Weights licence: Apache-2.0
- Licence evidence: https://github.com/zju3dv/LoFTR/blob/df7ca80f917334b94cfbe32cc2901e09a80e70a8/LICENSE ; https://huggingface.co/kornia/loftr/tree/0fb3b456cf176746709c38fa9cf80695a1c82dac
- Training data: MegaDepth (Internet photo collections with SfM / MVS depth) (loftr-outdoor: MegaDepth. loftr-indoor: ScanNet (ScanNet terms of use: non-commercial research), README training section.)
- Options:
long_side= 840 (>= 0) - Kernel libraries:
igev,cutlass_conv,cudnn_hacc - Variants:
loftr-outdoor,loftr-indoor - Not in this release (planned for 0.2.1):
loftr-outdoorONNX / TensorRT export: torch.export stops at a data-dependent branch (loftr.py:190);loftr-indoorONNX / TensorRT export: torch.export fails (TorchExportError, step 1/3 of the ONNX export) - Outputs: keypoints1, keypoints2, pairs, scores
- Input size: any H x W >= 32, padded to a multiple of 32
- Tested up to (portable, fp16, RTX 5060 Ti 16 GB):
loftr-outdoor3000x4000 (1.9 GiB peak allocated, 2.8 reserved);loftr-indoor3000x4000 (2.0 GiB peak allocated, 2.8 reserved) loftr-outdoor768x1024, first call: portable + torch.compile 33 s (cold Inductor cache).loftr-indoor768x1024, first call: portable + torch.compile 30 s (cold Inductor cache).loftr-outdoororiginal 768x1024: upstream LoFTR default_cfg, temp_bug_fix False, long side 840loftr-indoororiginal 768x1024: upstream LoFTR default_cfg, temp_bug_fix True, long side 840loftr-outdoorload / first call: original 0.35 s / 0.242 s, portable 0.24 s / 0.578 s, kernels 0.67 s / 0.369 sloftr-indoorload / first call: original 1.49 s / 0.693 s, portable 0.32 s / 0.53 s, kernels 0.54 s / 0.429 s- Protocol, first call, 3 warm-up, peak reset, median of 3 rounds x median of 5
Measured
Dense matchers: 5000 matches per pair sampled from the warp by certainty (upstream's balanced KDE sampling); sparse matchers: all their matches of one call (LightGlue, XFeat: LightGlue's MegaDepth-1500 setting; LoFTR, DeDoDe: their card defaults). OpenCV essential-matrix RANSAC (0.5 px threshold, normalised by the mean focal length); AUC of the larger of the rotation and translation-direction errors. Original = the original code named under Details, same checkpoint and protocol.
| variant | dataset | original fp32 | portable | portable fp32 | kernels | note |
|---|---|---|---|---|---|---|
loftr-outdoor | MegaDepth-1500, relative pose AUC@5 / @10 / @20 degrees | 50.1 / 67.5 / 80.3 | 50.6 / 67.6 / 80.3 | 50.8 / 67.8 / 80.2 | ||
loftr-outdoor | MegaDepth-1500, every 10th pair, relative pose AUC@5 / @10 / @20 degrees | 53.8 / 71.8 / 83.7 | 52.1 / 71.0 / 83.3 | 57.1 / 74.2 / 85.0 |
Measured on portable: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 3a7e4880 / aee4b565, 2026-11-13 to 2026-11-15; portable32: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 3a7e4880, 2026-11-15; kernels: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks 68b2f345, 2026-11-16.
Parity
CPU, float32: max |diff| <= 3e-05 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.0025 and p99 <= 0.02 (absolute error of the output keypoints1).
| variant | kernels error vs original, mean / p99 | GPU, torch | kernels used |
|---|---|---|---|
loftr-outdoor | 0.00211 / 0.01445 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, cutlass_conv@988138e042cb:sm_120, igev@f075827d68c6:sm_120 |
loftr-indoor | 0.00174 / 0.01007 | RTX 5060 Ti, torch 2.14.1+cu130 | cudnn_hacc@cudnn92400, cutlass_conv@988138e042cb:sm_120, igev@f075827d68c6:sm_120 |
Limitations
- Sparse output: no dense warp.
result.matches()gives the matched points forestimate_pose; every match is its own keypoint pair (pairsis[[0, 0], [1, 1], ...]). - One image pair per call internally (batches run pair by pair); the match counts differ per pair.
- No exportable module: the match counts are data-dependent (
torch.compileruns with graph breaks; ONNX / TensorRT export fails). - Images are converted to grayscale and matched at
long_side(default 840, upstream's MegaDepth test size); the coarse matching holds an (H W / 64)^2 score matrix, so memory grows with the 4th power of the side. The indoor weights were evaluated by upstream at 640 x 480 (ScanNet).
Example output
No example rendered for this checkpoint yet.