LocoTrack (fastforks)
LocoTrack (Cho et al., ECCV 2024): tracking any point offline over a whole video, TAPIR-like with local 4D correlation. A ResNet encodes every 256 x 256 frame at three resolutions; two cost volumes against the query's features give the initial position, visibility and uncertainty; then four refinements correlate the 7 x 7 neighbourhood of each track point with the 7 x 7 support of its query (a 4D correlation, encoded by small convs) and a transformer with ALiBi attention over the frames of each track corrects the whole track at once.
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/KU-CVLAB/LocoTrack at 3385a233a270.
Task: tracking. Tier: experimental. Outputs: tracks, visibility. Backends: portable, kernels.
Input: a video (its frames in order) and query points (t, x, y). Output: the position of every query in every frame and the probability that it is visible there.
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 |
|---|---|---|---|---|
locotrack-b@1 |
locotrack_base.ckpt | 8a5adbaeb610 |
Apache-2.0 | yes |
locotrack-s@1 |
locotrack_small.ckpt | da023594e6d6 |
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: Upstream README: trained on the panning Kubric MOVi-E sequences (synthetic, released as hamacojr/LocoTrack-panning-MOVi-E). No dataset entry here yet..
Licence evidence: https://huggingface.co/datasets/hamacojr/LocoTrack-pytorch-weights/blob/a8754d0edfd9e53713dcb31ee21734aac8b727cf/README.md (license: apache-2.0); https://github.com/KU-CVLAB/LocoTrack/blob/3385a233a270160b253a69050c6a3934527f7775/LICENSE (Apache-2.0)
Input size
any H x W >= 32, padded to a multiple of 32
Tested up to (portable, fp16, RTX 5060 Ti 16 GB): locotrack-b 3000x4000 (5.5 GiB peak allocated, 5.9 reserved); locotrack-s 3000x4000 (5.4 GiB peak allocated, 5.8 reserved)
Speed and memory
Tracking, synthetic, 480x854
| model | time (ms): original | portable | kernels | kernels speed-up: vs original | vs portable | peak memory (MiB, allocated / reserved): original | portable | kernels |
|---|---|---|---|---|---|---|---|---|
locotrack-b |
203 | 188 | 173 | 1.2x | 1.1x | 719 / 784 | 603 / 656 | 614 / 660 |
locotrack-s |
193 | 180 | 151 | 1.3x | 1.2x | 706 / 812 | 759 / 842 | 596 / 642 |
Startup and first call
| model | load (s): original | portable | kernels | first call (s): original | portable | kernels |
|---|---|---|---|---|---|---|
locotrack-b |
0.79 | 0.35 | 0.18 | 1.06 | 0.72 | 0.87 |
locotrack-s |
0.29 | 0.06 | 1.13 | 0.23 | 0.65 | 0.66 |
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) |
|---|---|---|---|---|---|---|---|---|---|---|---|
locotrack-b |
480x854 | 188 ms | 82.6 ms | planned for 0.2.1 | 173 ms | 146 s | β | 603 / 656 | 1117 / 1140 | 614 / 660 | β |
locotrack-s |
480x854 | 180 ms | 74.1 ms | planned for 0.2.1 | 151 ms | 140 s | β | 759 / 842 | 1000 / 1034 | 596 / 642 | 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
TAP-Vid protocol: frames resized to 256 x 256 (PIL Lanczos) before the model (AllTracker: the original frames, resized by the model), one query per track at its first visible point, tracked online (TAPNext, Track-On) or over the whole clip (TAPIR, AllTracker, LocoTrack); AJ, delta_avg (< 1, 2, 4, 8, 16 px at 256 x 256) and occlusion accuracy averaged over the videos. Original = the original code named under Details, same checkpoint and protocol.
| variant | dataset | original fp32 | original fp16 | portable | kernels | note |
|---|---|---|---|---|---|---|
locotrack-b |
TAP-Vid DAVIS (query first), every 3rd video, AJ / delta_avg / OA % | 67.19 / 78.05 / 89.58 | 67.31 / 78.11 / 89.85 | 67.31 / 78.07 / 89.89 | 67.29 / 78.10 / 89.83 |
Measured on NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks db9e79d5, 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| <= 2.3e-05 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.0036 and p99 <= 0.021 (absolute error of the output tracks).
Limitations
- Offline: the whole video is encoded before tracking (features of every frame in memory, about 7 MB per frame in float32); for live video use an online tracker (TAPNext, Track-On2).
- Frames are resized to 256 x 256 (the training size and that of upstream's TAP-Vid numbers), whatever their size and aspect ratio. Upstream's higher-resolution refinement is not ported.
- One video per call; all queries of a call share the video.
Usage
from fastforks import tracking
model = tracking.load("locotrack-b") # mode="auto": the preferred backend from recorded measurements
result = model(frames, [[0, x, y]]) # frames: the video's images in order; queries (t, x, y)
result.tracks # [T, Q, 2] float32: x, y in input pixels; result.visible [T, Q] bool
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: locotrack-b, locotrack-s.
Kernels backend
tracking.load("locotrack-b", mode="kernels") runs our CUDA kernels (fvm_conv). 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.