Models · Point tracking

CoTracker3

CoTracker3 (Meta AI, 2024): tracks any point through a video, all points jointly: a transformer refines the tracks with correlation features, attending over time and across tracks; offline (the whole video at once) and online (16-frame windows). Non-commercial: CC-BY-NC-4.0 code and weights.

experimental Fork of facebookresearch/co-tracker@82e02e8 · backends: portable, kernels

Checkpoint

Paper 2024 Added in 0.2.0 Evaluated with 0.2.0

Kernels, RTX 5060 Ti 16 GB 154 ms 480x854, fp16
vs the faster original 1.6× original fp16: 244 ms
Peak memory 757 MiB original: 2,221 MiB
Accuracy vs original fp32 +0.04 kernels aj, TAP-Vid DAVIS (query first) (65.02 original)

Use it

pip install fastforks

Lean install with only what this model needs: pip install "fastforks-core[nc]"

from fastforks import tracking

model = tracking.load("cotracker3-offline")  # kernels on a supported GPU, else portable PyTorch
result = model(video, queries)
result.tracks  # also: visibility
result.backend  # what ran: mode, device, dtype, kernels

The weights are non-commercial: fastforks downloads them only with FASTFORKS_NONCOMMERCIAL=1 set.

fastforks bench cotracker3-offline

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, 480x854 (the task’s standard size)

BackenddtypeTime per call (ms)vs original fp16Peak memory
allocated / reserved
Load (s)First call (s)
Original fp32fp32not measured ––––
Original fp16the authors' mixed-precision option (float16); facebookresearch/co-tracker@82e02e802975 as its predictor, offline (whole video, 6 x 6 support grid), frames resized to 384 x 512fp16 244–2,221 MiB/2,480 MiB0.30.32
Portable fp32fp32not measured ––––
Portablefp16 1621.5×755 MiB/882 MiB0.20.47
Kernelsfp16 1541.6×757 MiB/882 MiB0.30.49
Portable + torch.compile 1351.8×755 MiB/902 MiB36 (build)–
TensorRT fp16fp16 25.99.4×284 MiBenginenot comparable157 (build)–

Measured on RTX 5060 Ti 16 GB, torch 2.14.1+cu130, 2026-11-14 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

DeviceBackenddtypems / callPeak memoryOutput vs referenceInput
Apple M1 Pro 16 GB MPS, torch 2.14.1, 2026-11-17Portablefp321430 4,725 MiB processskip480x854
Kernelsfp16not implemented–480x854

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

Non-commercial only

Code
CC-BY-NC-4.0 LICENSE.md
Weights
CC-BY-NC-4.0 source
Training data
Upstream paper: trained on synthetic Kubric MOVi-F sequences, then fine-tuned by pseudo-labelling about 15k real internet videos with other trackers (not released). No dataset entry here yet.
  • code: CC-BY-NC-4.0 (noncommercial)
  • weights: CC-BY-NC-4.0

Our reading of the licences, not legal advice.

Weights

scaled_offline.pth 102 MB· hosted on Hugging Face

fastforks downloads and checks these files on first use (fastforks download cotracker3-offline). Non-commercial weights need FASTFORKS_NONCOMMERCIAL=1.

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.

TAP-Vid DAVIS (query first)

every 3rd video, AJ / delta_avg / OA %

Rundtypeaj ↑delta_avg ↑oa ↑Δ aj vs fp32 originalvideosms / video
Original fp32fp3265.0278.7588.24 reference 10 –
Original fp16fp1665.0878.7888.27 +0.06 10 –
Portable fp32fp3265.0478.7588.26 +0.02 10 –
Portablefp1665.0778.7588.26 +0.05 10 –
Kernelsfp1665.0678.7288.27 +0.04 10 –

Query-first 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 authors’ code, same checkpoint and protocol. Measured on original fp32, original fp16: RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-14; portable fp32: RTX 5060 Ti, torch 2.14.1+cu130, fastforks bb4e8b64, 2026-11-17; portable, kernels: RTX 5060 Ti, torch 2.14.1+cu130, fastforks f104019b, 2026-11-14.

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| <= 6e-05 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.0027 and p99 <= 0.014 (absolute error of the output tracks).

CheckToleranceMeasured
CPU fp32, max |diff|≤ 0.00006 not recorded
GPU default dtypes, mean |diff|≤ 0.0027 0.00177
GPU default dtypes, p99 |diff|≤ 0.014 0.00952

Recorded for cotracker3-offline on RTX 5060 Ti, torch 2.14.1+cu130, 2026-11-15, with these kernels: fvm_conv

Details

CoTracker3 (Karaev et al., Meta AI 2024): tracking any point with a transformer that tracks all points jointly. A CNN encodes every frame at 384 x 512 into 128-channel features at 1/4 resolution (a 4-level pyramid); each track correlates the 7 x 7 neighbourhood of its current position with the 7 x 7 support of its query on every level (an MLP encodes the 4D correlation), and a transformer with attention over time and, through 64 virtual tracks, across tracks updates positions, visibility and confidence in 6 iterations. Offline: the whole video at once; online: sliding windows of 16 frames, 8 new frames per step.

Input / output
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.
Input size
any H x W >= 32, padded to a multiple of 32; tested up to 3000x4000 (5,615 MiB allocated)
Outputs
tracks, visibility
Checkpoint
online=false, window_len=60
CUDA kernels
  • fvm_convfused convolutions for image classifiers
Apple (MPS)
fp32

Limitations

  • Frames are resized to 384 x 512 (the training size and that of upstream's predictor), whatever their size and aspect ratio.
  • cotracker3-offline holds the features of every frame (about 4 MB per frame in float16); cotracker3-online processes 16-frame windows and only keeps the tracks.
  • Visibility is upstream's score, rescaled so that 0.5 is upstream's threshold: the offline predictor's visibility > 0.9, the online predictor's visibility x confidence > 0.6.
  • One video per call; all queries of a call share the video.
  • cotracker3-online: the online window loop is data-dependent (6 graph breaks; `torch.export` guards on data), so no ONNX / TensorRT export; cotracker3-offline exports.
Full reference, generated from the repository

CoTracker3 (Meta AI, 2024): tracks any point through a video, all points jointly: a transformer refines the tracks with correlation features, attending over time and across tracks; offline (the whole video at once) and online (16-frame windows). Non-commercial: CC-BY-NC-4.0 code and weights.

Commercial use: no (code: CC-BY-NC-4.0 (noncommercial); weights: CC-BY-NC-4.0).

variantinputbackendtime (ms)peak memory (MiB, allocated / reserved)
cotracker3-offlinesynthetic (480x854)original
portable
kernels
portable + torch.compile
TensorRT fp16
244
162
154
135
25.9
2221 / 2480
755 / 882
757 / 882
755 / 902
519 / 586
cotracker3-onlinesynthetic (480x854)original
portable
kernels
portable + torch.compile
TensorRT fp16
328
184
176
159
planned for 0.2.1
1659 / 1886
755 / 906
757 / 888
701 / 846
–
Details
  • Original code: https://github.com/facebookresearch/co-tracker at 82e02e802975
  • Code licence: CC-BY-NC-4.0 (noncommercial); files: LICENSE.md
  • Weights licence: CC-BY-NC-4.0
  • Licence evidence: https://github.com/facebookresearch/co-tracker/blob/82e02e8029753ad4ef13cf06be7f4fc5facdda4d/LICENSE.md (CC BY-NC 4.0); https://github.com/facebookresearch/co-tracker/blob/82e02e8029753ad4ef13cf06be7f4fc5facdda4d/README.md#license; https://huggingface.co/facebook/cotracker3/blob/bf55ea50d4390e1820a267f131cd6587240fb2c5/README.md (license: cc-by-nc-4.0)
  • Training data: Upstream paper: trained on synthetic Kubric MOVi-F sequences, then fine-tuned by pseudo-labelling about 15k real internet videos with other trackers (not released). No dataset entry here yet.
  • Kernel libraries: fvm_conv
  • Variants: cotracker3-offline, cotracker3-online
  • Not in this release (planned for 0.2.1): cotracker3-online ONNX / TensorRT export: torch.export fails (TorchExportError, step 1/3 of the ONNX export)
  • Outputs: tracks, visibility
  • Input size: any H x W >= 32, padded to a multiple of 32
  • Tested up to (portable, fp16, RTX 5060 Ti 16 GB): cotracker3-offline 3000x4000 (5.5 GiB peak allocated, 5.9 reserved); cotracker3-online 3000x4000 (5.5 GiB peak allocated, 5.9 reserved)
  • cotracker3-offline 480x854, first call: portable + torch.compile 36 s (cold Inductor cache), TensorRT fp16 157 s (ONNX + engine).
  • cotracker3-online 480x854, first call: portable + torch.compile 38 s (cold Inductor cache).
  • cotracker3-offline original 480x854: the authors' mixed-precision option (float16); facebookresearch/co-tracker@82e02e802975 as its predictor, offline (whole video, 6 x 6 support grid), frames resized to 384 x 512
  • cotracker3-online original 480x854: the authors' mixed-precision option (float16); facebookresearch/co-tracker@82e02e802975 as its predictor, online (16-frame windows, step 8), frames resized to 384 x 512
  • cotracker3-offline load / first call: original 0.65 s / 0.866 s, portable 0.28 s / 0.416 s, kernels 0.19 s / 0.52 s
  • cotracker3-online load / first call: original 0.54 s / 0.536 s, portable 0.27 s / 0.472 s, kernels 0.18 s / 0.515 s
  • Protocol, first call, 3 warm-up, peak reset, median of 3 rounds x median of 5

Measured

Query-first 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.

variantdatasetoriginal fp32original fp16portableportable fp32kernelsnote
cotracker3-offlineTAP-Vid DAVIS (query first), every 3rd video, AJ / delta_avg / OA %65.02 / 78.75 / 88.2465.08 / 78.78 / 88.2765.07 / 78.75 / 88.2665.04 / 78.75 / 88.2665.06 / 78.72 / 88.27
cotracker3-onlineTAP-Vid DAVIS (query first), every 3rd video, AJ / delta_avg / OA %70.34 / 80.63 / 94.2870.28 / 80.62 / 94.2670.32 / 80.62 / 94.2670.34 / 80.64 / 94.3070.25 / 80.63 / 94.22

Measured on portable: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks f104019b, 2026-11-14; portable32: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks bb4e8b64, 2026-11-17; kernels: NVIDIA GeForce RTX 5060 Ti, torch 2.14.1+cu130, fastforks f104019b, 2026-11-14.

Parity

CPU, float32: max |diff| <= 6e-05 vs the original. GPU, float16 (the default), portable and kernels: mean |diff| <= 0.0027 and p99 <= 0.014 (absolute error of the output tracks).

variantkernels error vs original, mean / p99GPU, torchkernels used
cotracker3-offline0.00177 / 0.00952RTX 5060 Ti, torch 2.14.1+cu130fvm_conv@f1d064e539e1:sm_120
cotracker3-online0.00149 / 0.00738RTX 5060 Ti, torch 2.14.1+cu130fvm_conv@f1d064e539e1:sm_120

Limitations

  • Frames are resized to 384 x 512 (the training size and that of upstream's predictor), whatever their size and aspect ratio.
  • cotracker3-offline holds the features of every frame (about 4 MB per frame in float16); cotracker3-online processes 16-frame windows and only keeps the tracks.
  • Visibility is upstream's score, rescaled so that 0.5 is upstream's threshold: the offline predictor's visibility > 0.9, the online predictor's visibility x confidence > 0.6.
  • One video per call; all queries of a call share the video.
  • cotracker3-online: the online window loop is data-dependent (6 graph breaks; torch.export guards on data), so no ONNX / TensorRT export; cotracker3-offline exports.

Example output

No example rendered for this checkpoint yet.