Feiyu Chen
Feiyu Chen
I think you can plug in the deep sort algorithm into this method: https://github.com/felixchenfy/Realtime-Action-Recognition/blob/master/utils/lib_tracker.py#L42
从人的骨架提取位置特征+机器学习，见README第一节：https://github.com/felixchenfy/Realtime-Action-Recognition#1-algorithm
hi，抱歉除了README的介绍以外就没有啦。 我当时没有看什么文献 --- 这个项目只是应用了一下机器学习来做分类
不了吧😂 我自毕业以来就没做过机器学习了，水平不行。如果真要探讨，我们邮件讨论吧，[email protected]，谢谢
流程/算法我写在README里了啊： https://github.com/felixchenfy/Realtime-Action-Recognition#1-algorithm https://github.com/felixchenfy/Data-Storage/blob/master/EECS-433-Pattern-Recognition/FeiyuChen_Report_EECS433.pdf
这不是一言两语能说的清的啊... 得一步步debug
Sorry I don't have a quick answer now, since it's been many years since I worked on it. Thanks for reporting the issue anyway
- I'm not sure... I think the training shouldn't be slow even using CPU, because the features used for training are human's positions, whose feature dimensionality is pretty small. It...
Hello, the code you are asking is from this package "https://github.com/ildoonet/tf-pose-estimation". You might get more answers if the question is asked there.
@rahul-moorkoth Hi, thanks for reporting this issue!