A-contrario Smooth TRajectory Extraction


Download / Installation and usage manual

Snow data sequence


We filmed moving snowflakes in front of a tainted wall at a high frequency (210Hz), and the points were detected using very simple means (thresholding and a automatic clean-up, see our accompanying paper) and the ground-truth was then recovered by hand on this simple to track version. The snow sequence has then been subsampled to 30Hz by keeping one frame out of every 7, to obtain a challenging to track point clouds sequence.

Frames 35, 36 and 37 of the (30Hz) sequence with the corresponding detections.
Snow sequence. The snow flakes have trajectories of varying smoothness and speed, while the taints on the background wall stay in place and only have some jitter motion. The superposition of two different kinds of motion make this sequence interesting to use for algorithm comparison.



Feel free to use this sequence in your publications, but please cite the accompanying paper as a reference to its origin.

  • Download the snow sequence (100 kB)
    Snow sequence containing the points and the ground truth trajectories, and the results of the ASTRE algorithm

ASTRE results

Ground truth.
Detections obtained using the ASTRE algorithm (with holes), for the maximal log(NFA) parameter log10(ε) = 5.

Here are the results for each algorithm (astre-noholes and astre-holes) for the standard parameter log10(ε) = 0 and the optimal parameter log10(ε) = 5 in terms of recall and precision (computed on the number of correct links found) and the number of found trajectories as a reference:

algorithm recall precision # trajs
astre-noholes, log10(ε) = 0 0.50 0.99 39
astre-holes, log10(ε) = 0 0.55 0.91 34
astre-noholes, log10(ε) = 5 0.71 0.95 124
astre-holes, log10(ε) = 5 0.76 0.92 91