【多人姿态估计】AlphaPose_yolov4推理demo复现

Myth丶恋晨 2022-10-22 13:51 181阅读 0赞

代码：[https://github.com/WildflowerSchools/AlphaPose][https_github.com_WildflowerSchools_AlphaPose]

此处环境与yolov3不同，注意。

# 1.环境 #

cuda101
    
    torch==1.6.0
    torchvision==0.7.0
    opencv-python==4.1.0.25
    tqdm
    natsort
    Cython
    pycocotools

python setup.py build develop --user
    apt-get update
    apt-get -y install python3.6-tk

对于python setup.py build develop --user，python库安装过程中慢的，使用：

pip --default-time=1000 install -i https://pypi.tuna.tsinghua.edu.cn/simple ***

# 2.模型准备与数据准备 #

参考：[https://blog.csdn.net/qq\_35975447/article/details/114654804][https_blog.csdn.net_qq_35975447_article_details_114654804]

不一样的地方就是下载的是[yolov4.weigts][]。或者此处：[https://github.com/AlexeyAB/darknet][https_github.com_AlexeyAB_darknet]

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM1OTc1NDQ3_size_16_color_FFFFFF_t_70][]

# 3.demo--跟踪 #

python scripts/demo_inference.py --cfg configs/coco/resnet/256x192_res50_lr1e-3_2x-dcn.yaml --checkpoint pretrained_models/fast_dcn_res50_256x192.pth --indir examples/demo/ --vis --showbox --save_img --pose_track --sp

![watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM1OTc1NDQ3_size_16_color_FFFFFF_t_70 1][]

跟踪轨迹可以参考deep-sort-yolov3中画轨迹。或者参考[https://xiaorun.blog.csdn.net/article/details/80572523][https_xiaorun.blog.csdn.net_article_details_80572523]

跟踪哪个关节点，就将对应的关节点坐标加入到deque中。

keypoints中的0-17分别代表
    
    0 noise
    1 left_eye
    2 right_eye
    3 left_ear
    4 right_ear
    5 left_shoulder
    6 right_shoulder
    7 left_elbow
    8 right_elbow
    9 left_wrist
    10 right_wrist
    11 left_hip
    12 right_hip
    13 left_knee
    14 right_knee
    15 left_ankle
    16 right_ankle

此处参考：[https://zhuanlan.zhihu.com/p/121452714][https_zhuanlan.zhihu.com_p_121452714]

[https_github.com_WildflowerSchools_AlphaPose]: https://github.com/WildflowerSchools/AlphaPose
[https_blog.csdn.net_qq_35975447_article_details_114654804]: https://blog.csdn.net/qq_35975447/article/details/114654804
[yolov4.weigts]: https://blog.csdn.net/jy1023408440/article/details/105736611?ops_request_misc=%257B%2522request%255Fid%2522%253A%2522161596930916780271531555%2522%252C%2522scm%2522%253A%252220140713.130102334..%2522%257D&request_id=161596930916780271531555&biz_id=0&utm_medium=distribute.pc_search_result.none-task-blog-2~all~sobaiduend~default-1-105736611.first_rank_v2_pc_rank_v29_10&utm_term=yolov4.weight
[https_github.com_AlexeyAB_darknet]: https://github.com/AlexeyAB/darknet
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM1OTc1NDQ3_size_16_color_FFFFFF_t_70]: /images/20221022/705d1c19956747049d670f751737d1dd.png
[watermark_type_ZmFuZ3poZW5naGVpdGk_shadow_10_text_aHR0cHM6Ly9ibG9nLmNzZG4ubmV0L3FxXzM1OTc1NDQ3_size_16_color_FFFFFF_t_70 1]: /images/20221022/468808870eff49fdb0524febe9910c40.png
[https_xiaorun.blog.csdn.net_article_details_80572523]: https://xiaorun.blog.csdn.net/article/details/80572523
[https_zhuanlan.zhihu.com_p_121452714]: https://zhuanlan.zhihu.com/p/121452714