Track manipulation, locomotion and embodied systems.
Search papers, labs, and topics across Lattice.
This reproducibility study rigorously audits an ICPR 2026 pipeline that fine-tunes Video Vision Transformers (ViViT) with focal loss and Taguchi L18 augmentations to identify dangerous football tackles. The authors fully reproduce the reported headline metrics (0.667 risky recall, 0.588 F1) and release an execution artifact compliant with student athlete privacy constraints. Crucially, the ablation reveals that brightness attenuation is the dominant factor driving performance, and unaugmented ViViT models drop to 0.545 recall—failing to outperform a standard C3D baseline (0.583).
Video Vision Transformers collapse below legacy C3D baselines on safety-critical action detection unless propped up by photometric augmentation, where simple brightness reduction surprisingly drives nearly all performance gains over spatial transforms.
This paper is a Track 2 reproducibility companion to an ICPR 2026 study on risky tackle detection in American football prac- tice videos. The original work fine-tuned a Video Vision Transformer (ViViT) on 733 clips labeled with the SATT-3 rubric. It used focal loss, Taguchi L18 augmentation, and 5-fold cross-validation. It reported risky- class recall of 0.67 and risky-class F1 of 0.59. This companion documents the released artifact and traces those numbers to specific scripts, fold out- puts, and aggregation files. The reproduced headline is run_15. It com- bines Gaussian noise with static brightness decrease and uses no rotation and no flip. Its fold-mean risky recall is 0.667 and its fold-mean risky F1 is 0.588. These values match the published headline after rounding. The ablation shows that brightness is the dominant factor. Its risky-recall main-effect range is 0.055, which is larger than the ranges for rotation, flip, and noise. Without augmentation, ViViT reaches risky recall of 0.545 and does not exceed the C3D baseline of 0.583. The raw clips show iden- tifiable student athletes, so they cannot be redistributed. The artifact provides a public sample for pipeline checks and a controlled route for full-data review.