Dataset: Deep learning-enhanced PTV tracking of rod-like particles in non-Newtonian fluids — trajectory data and YOLOv11 model weights

This dataset contains filtered particle trajectory data supporting the findings of Wolff et al. (2026), submitted to Measurement Science and Technology. It includes tracked fiber trajectories for five fiber loadings (25–800 fibers) of steel fibers in Carbopol yield-stress gel, acquired at 200 fps with a spatial resolution of 0.140 mm/pixel. Trajectories were reconstructed using a YOLOv11 instance segmentation model combined with an α-β-γ predictive filter. Tracks shorter than 20 frames have been removed in post-processing. See README.md for full field descriptions and units.