V-JEPA 2 Meets AQA: Phase-Based Scoring and Reference-Based Feedback for Mountain Biking
Computer Science > Computer Vision and Pattern Recognition arXiv:2610.30160 (cs) [Submitted on 27 Sep 2026] Title: V-JEPA 2 Meets AQA: Phase-Based Scoring and Reference-Based Feedback for Mountain Biking Authors: Santhiya Rajan, Savio Cardoz.
计算机科学 > 计算机视觉与模式识别 arXiv:2610.30160 (cs) [提交于 2026 年 9 月 27 日] 标题：V-JEPA 2 遇见 AQA：山地自行车运动中基于阶段的评分与基于参考的反馈 作者：Santhiya Rajan, Savio Cardoz。
Abstract: We study mountain-bike action quality assessment (AQA) with weak labels and candidate feedback. The training corpus comprises 95 field-recorded jump clips with binary labels denoting good or poor technique. These are coarse categorical labels, not continuous expert quality scores; phase-boundary and aspect-level annotations are unavailable.
摘要：我们研究了带有弱标签和候选反馈的山地自行车动作质量评估（AQA）。训练语料库包含 95 段实地录制的跳跃视频片段，并带有表示技术好坏的二元标签。这些是粗略的分类标签，而非连续的专家质量评分；且缺乏阶段边界和方面级别的标注。
Our pipeline summarizes four heuristic phases using frozen V-JEPA 2 video features and two-dimensional pose and motion measurements. A small transformer predicts a binary-label score. A separate reference layer compares phase-specific measurements with those of clips labelled as good technique to propose phase, body-parameter, and direction cues.
我们的流程利用冻结的 V-JEPA 2 视频特征以及二维姿态和运动测量数据，总结了四个启发式阶段。一个小型 Transformer 模型用于预测二元标签分数。一个独立的参考层将特定阶段的测量结果与标记为良好技术的片段进行比较，从而提出阶段、身体参数和方向提示。
In five-fold clip-level cross-validation, the full scorer obtained pooled out-of-fold AUROC 0.768 and average precision 0.812. These are exploratory figures because the reporting fold also selected the neural training epoch and riders were not separated between folds. Removing biomechanical scoring inputs raised AUROC to 0.809.
在五折片段级交叉验证中，完整评分器获得的汇总折外 AUROC 为 0.768，平均精度为 0.812。这些数据属于探索性数值，因为报告折同时也选择了神经网络训练周期，且骑手在各折之间未进行区分。移除生物力学评分输入后，AUROC 提升至 0.809。
On a 17-clip external probe, an earlier checkpoint obtained binary accuracy 0.412 and Spearman association -0.135 with expert scores. Candidate cues had precision 0.118 and recall 1.0 against each of two coaches, with substantial disagreement about direction. These findings do not establish rider-independent scoring or reliable corrective feedback; they identify the empirical work needed to evaluate either use.
在 17 段视频的外部探测中，一个早期检查点获得的二元准确率为 0.412，与专家评分的斯皮尔曼相关系数为 -0.135。候选提示在对比两位教练时，精确率为 0.118，召回率为 1.0，且在方向上存在显著分歧。这些发现并未建立起独立于骑手的评分系统或可靠的纠正反馈；它们指出了评估上述任一用途所需的实证工作。