用迁移学习提升瑜伽动作识别准确率，最高达96%。
Yoga Pose Classification Using Transfer Learning
- 基于预训练模型微调，结合神经架构搜索优化结构
- DenseNet-121在Yoga-82数据集上达到85%精度（Top-1）
- 适合对姿态识别与健康应用感兴趣的开发者
瑜伽已成为维持身心健康的重要方式。随着生活节奏加快、居家办公增多，人们难以投入时间去健身房锻炼，因此人体姿态估计成为关键问题，需精确定位身体关键点。针对大规模瑜伽动作识别的挑战性数据集Yoga-82（含82个类别），本文采用VGG-16、ResNet-50、ResNet-101和DenseNet-121等预训练模型，并通过不同方式微调以提升性能。同时引入神经架构搜索，在预训练架构基础上增加层数以优化表现。实验结果显示，DenseNet-121取得最佳效果：Top-1准确率为85%，Top-5准确率达96%，优于当前最先进水平。
原文摘要 · Abstract (English)
Yoga has recently become an essential aspect of human existence for maintaining a healthy body and mind. People find it tough to devote time to the gym for workouts as their lives get more hectic and they work from home. This kind of human pose estimation is one of the notable problems as it has to deal with locating body key points or joints. Yoga-82, a benchmark dataset for large-scale yoga pose recognition with 82 classes, has challenging positions that could make precise annotations impossible. We have used VGG-16, ResNet-50, ResNet-101, and DenseNet-121 and finetuned them in different ways to get better results. We also used Neural Architecture Search to add more layers on top of this pre-trained architecture. The experimental result shows the best performance of DenseNet-121 having the top-1 accuracy of 85% and top-5 accuracy of 96% outperforming the current state-of-the-art result.
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。