Simple Baselines for Human Pose Estimation and Tracking 论文解读

布满荆棘的人生 2024-04-19 14:29 79阅读 0赞

**paper title**: simple baselines for human pose estimation and tracking  
**paper link**: https://arxiv.org/abs/1804.06208  
**github**:https://github.com/microsoft/human-pose-estimation.pytorch  
**conf & anthor**: ECCV 18,Bin Xiao et al. (MSRA,UEST)  
**arXiv submit v1**: 2018.04

姿态估计系列：

*  [DeepPose - human pose estimation via deep nerual networks][]
 *  [Efficient Object Localization Using Convolutional Networks][]
 *  [Convolutional Pose Machines ][Convolutional Pose Machines]
 *  [Human Pose Estimation with Iterative Error Feedback][]
 *  [Hourglass Network - stacked hourglass network for human pose estimation ][Hourglass Network - stacked hourglass network for human pose estimation]
 *  [OpenPose - Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields][]
 *  [RMPE: Regional Multi-person Pose Estimation][RMPE_ Regional Multi-person Pose Estimation]
 *  [Pose flow: Efficient Online Pose Tracking][Pose flow_ Efficient Online Pose Tracking]
 *  [CrowdPose: Efficient Crowded Scenes Pose Estimation and A New Benchmark][CrowdPose_ Efficient Crowded Scenes Pose Estimation and A New Benchmark]

#### 主要内容： ####

这篇文章提出了一种简洁高效的姿态估计和跟踪baseline方法，姿态估计方面使用ResNet+deconvolution, 跟踪方面使用optical Flow(光流)。

#### 网络结构： ####

**1.姿态估计**  
使用Faster-RCNN生成人的proposal，然后在关键点检测部分，使用ResNet作为主干网络，然后接三个反卷积层，生成最后的heatmap。

**2.姿态跟踪**  
姿态跟踪方面使用基于Optical flow的方法。  
`1.Joint propagation using optical flow`  
对于前一帧中的检测到的人体各个关节点，计算optical flow移动后的值，然后确定一个boundingbox, 作为传播到当前帧中的人体proposal，这样可以弥补在运动模糊等情况下造成的人体proposal提取失败的情况。

`2.Flow-based pose similarity`  
帧间追踪需要计算不同“人”的相似度，使用boundingbox IoU在运动较快或者密集场景下效果不好，使用关节点的OKS(Object Keypoint Similarity)当两帧间的姿态相差较大效果也不好，都有一些弊端。这篇文章就利用了光流，先计算下一帧的加上光流偏移后的姿态，然后再计算OKS，考虑到两帧之间可能有各种原因造成的不连续情况，所以这里可以计算连续多帧的OKS。

`3.Flow-based pose tracking`  
首先将1中说的Joint propagation得到proposal和检测模型得到的proposal做NMS，然后进行单人姿态估计，之后计算如2中所述的两帧之间的OKS，并进行贪婪匹配，完成跟踪。

#### 实验和结果： ####

部分比较结果如下(图片来自原论文)：  
![simple baseline experiment result][]

这篇写的也有真点 `simple` 了 ╮（╯＿╰）╭

[DeepPose - human pose estimation via deep nerual networks]: https://blog.csdn.net/JerryZhang__/article/details/97561493
[Efficient Object Localization Using Convolutional Networks]: https://blog.csdn.net/JerryZhang__/article/details/97561974
[Convolutional Pose Machines]: https://blog.csdn.net/JerryZhang__/article/details/97562875
[Human Pose Estimation with Iterative Error Feedback]: https://blog.csdn.net/JerryZhang__/article/details/97563255
[Hourglass Network - stacked hourglass network for human pose estimation]: https://blog.csdn.net/JerryZhang__/article/details/98308729
[OpenPose - Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields]: https://blog.csdn.net/JerryZhang__/article/details/98667549
[RMPE_ Regional Multi-person Pose Estimation]: https://blog.csdn.net/JerryZhang__/article/details/100767171
[Pose flow_ Efficient Online Pose Tracking]: https://blog.csdn.net/JerryZhang__/article/details/100835787
[CrowdPose_ Efficient Crowded Scenes Pose Estimation and A New Benchmark]: https://blog.csdn.net/JerryZhang__/article/details/100937176
[simple baseline experiment result]: https://imgconvert.csdnimg.cn/aHR0cHM6Ly9ub3RlLnlvdWRhby5jb20veXdzL3B1YmxpYy9yZXNvdXJjZS9hMGQyM2ViZjE3ZmRmN2NlZGQ1YjYxNDkxZThiMWYzMC94bWxub3RlL1dFQlJFU09VUkNFYTJhNGM3YzExMjM2YzJlODJiZDg4NzkxMjFiMWQ3OGQvMjE2ODU?x-oss-process=image/format,png