Stacked Hourglass Networks for Human Pose Estimation

用于人体姿态估计的堆叠沙漏网络.

DeBERTa: Decoding-enhanced BERT with Disentangled Attention

DeBERTa：使用分解注意力机制和增强型掩膜解码器改进预训练语言模型.

DeepPose: Human Pose Estimation via Deep Neural Networks

DeepPose：通过深度神经网络实现人体姿态估计.

When Do You Need Billions of Words of Pretraining Data?

什么时候需要数十亿单词的预训练数据？

FCOS: A Simple and Strong Anchor-free Object Detector

FCOS：一种简单强力的Anchor-free目标检测器.

AMASS: Archive of Motion Capture as Surface Shapes

AMASS：经过SMPL参数标准化的三维人体动作捕捉数据集合.