数据集组合比数据量更重要，速度差异是改善帕金森步态评估的关键。
More Motion Is Not Always Better Motion: Corpus Composition Governs Whether Augmentation Helps SMPL-Based Parkinsonian Gait Severity Estimation
- 用六个不同任务组合的体感数据集训练编码器，对比其性能差异。
- 含速度变化的数据集最高得分为0.53，低于真实数据集0.58。
- 合成数据和单目视频无法提升效果，说明数据质量比数量重要。
我们使用三个冻结的MotionAGFormer编码器作为特征提取器，基于SMPL运动数据对MDS-UPDRS步态严重程度进行评分，在一个隐藏的多中心测试集上达到宏平均F1为0.58。由于系统成员仅在所用运动语料库上不同，单独评估各编码器可分离出语料库的贡献。从单一惯性数据集中选取的六组数据集，仅因包含的任务类型不同而有所差异，得分在0.32至0.53之间；其中仅一组优于未引入外部运动数据的编码器（0.51）。决定性能差异的关键并非数据量大小，而是是否包含步行速度的对比变化——该表征似乎依赖于此。进一步加入第三个采集站点但保持任务组成不变的数据集表现更差。同样，精确合成运动与单目重建网络视频均无法带来提升。尝试修改学习到的表示本身，反而使所有变体性能下降。
原文摘要 · Abstract (English)
We grade MDS-UPDRS gait severity from SMPL motion using three frozen MotionAGFormer encoders as featurizers, reaching macro-F1 0.58 on a hidden, multi-site test set. Because the system's members differ only in their lifting corpus, evaluating encoders singly on that test set isolates what that corpus contributes. Six pools drawn from one inertial dataset, varying only in which walking tasks they include, score between 0.32 and 0.53, and just one of them beats the 0.51 of an encoder given no outside motion at all. What separates them is not how much data they hold but whether they carry a contrast in walking speed, the variation this representation appears to depend : a further pool adding a third collection site at fixed task composition does worse still. The same rule explains why exact synthetic motion and monocularly reconstructed web video both fail to help. Modifying the learned representation itself, rather than the corpus behind it, cost every variant that attempted it.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。