DeDoDe v2 (fastforks)

DeDoDe v2 (Edstedt et al., 3DV 2024; detector v2: CVPR Workshops 2024): detect, don't describe, describe, don't detect. A VGG19 network with convolutional refiners finds 10000 keypoints per image (the detector, retrained in v2), a separate descriptor network (VGG19, or VGG19 + a frozen DINOv2 ViT-L/14 for -G) describes them, and the descriptions are matched as mutual maxima of a dual softmax. Returns both keypoint sets, the index pairs of the matches and their probability.

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/Parskatt/DeDoDe at 6d156183f4dc.

Task: matching. Tier: experimental. Outputs: keypoints1, keypoints2, pairs, scores. Backends: portable, kernels.

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. 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.

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
dedode-v2-b@1 dedode_detector_L_v2.pth
dedode_descriptor_B.pth
4113809dd9e0
8eccfc270ec9
MIT yes
dedode-v2-g@1 dedode_detector_L_v2.pth
dedode_descriptor_G.pth
dinov2_vitl14_pretrain.pth
4113809dd9e0
ef6e3f2911bb
d5383ea8f487
MIT yes

Licence

Original-code strings, recorded as they are (research-use package; check them before any other use). Code: MIT. Licence files: LICENSE, LICENSE-DINOv2.

Training data: MegaDepth (Internet photo collections with SfM / MVS depth) (Detector v2 and descriptors: MegaDepth (README training section, papers); DINOv2 (-G, frozen): LVD-142M (Meta).).

Licence evidence: https://github.com/Parskatt/DeDoDe/blob/6d156183f4dc84cd704ae779eebc8350995c5b06/README.md#license ; https://github.com/Parskatt/DeDoDe/blob/6d156183f4dc84cd704ae779eebc8350995c5b06/LICENSE

Input size

any H x W >= 32, padded to a multiple of 32 Tested up to (portable, fp16, RTX 5060 Ti 16 GB): dedode-v2-b 3000x4000 (1.9 GiB peak allocated, 3.8 reserved); dedode-v2-g 3000x4000 (2.5 GiB peak allocated, 4.5 reserved)

Speed and memory

Matching, synthetic, 768x1024

model time (ms): original fp32 original portable kernels kernels speed-up: vs original vs portable peak memory (MiB, allocated / reserved): original portable kernels
dedode-v2-b – 318 246 – – – 1734 / 1992 1535 / 1654 –
dedode-v2-g – 666 592 – – – 2335 / 2472 2142 / 2334 –

Startup and first call

model load (s): original portable kernels first call (s): original portable kernels
dedode-v2-b 3.73 0.67 – 1.10 0.75 –
dedode-v2-g 6.36 1.12 – 1.51 1.04 –

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)
dedode-v2-b 768x1024 246 ms 164 ms planned for 0.2.1 – 50.5 s – 1535 / 1654 1893 / 2434 – –
dedode-v2-g 768x1024 592 ms 330 ms planned for 0.2.1 – 86.4 s – 2142 / 2334 2881 / 3316 – –

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

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 fp16 portable note
dedode-v2-b MegaDepth-1500, every 10th pair, relative pose AUC@5 / @10 / @20 degrees 57.3 / 71.5 / 82.5 54.6 / 71.8 / 83.1

Measured on NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks aee4b565, 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| <= 0.011 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.002 and p99 <= 0.001 (absolute error of the output keypoints1).

Limitations

  • Sparse output: no dense warp. result.matches() gives the matched points for estimate_pose; pairs are the index pairs pycolmap / COLMAP databases take.
  • 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 resized to a size x size square (default 784, as upstream) whatever their aspect ratio; the dual softmax holds a max_keypoints^2 matrix (10000 keypoints: 400 MB per float32 copy).
  • Upright only: the released descriptors are not trained for in-plane rotations.

Usage

from fastforks import matching

model = matching.load("dedode-v2-b")  # mode="auto": the preferred backend from recorded measurements
result = model("view1.jpg", "view2.jpg")
result.keypoints1  # [2, H1, W1] float32: x, y in view2 pixels
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: dedode-v2-b, dedode-v2-g.

Kernels backend

matching.load("dedode-v2-b", mode="kernels") runs our CUDA kernels (igev, cutlass_conv, 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.

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. πŸ™‹ Ask for provider support

Collection including fastforks/dedode