Jiashuo

感谢您提供的解答！ 后期应该是处理视频流，关于视频行为识别等图像处理技术目前应该都有解决方案，我困惑的是如何在识别后，当我读取视频流的时候能判断当前数据块是视频对应的区域，从而做进一步的处理。 目前我所了解的大部分是处理完后再图像上作出标记，以供人眼识别，那么这些标记反映到底层数据包该怎么识别呢？ 您在“1”中提到的滑动窗口匹配应该是一种解决办法，但是对于动态视频来说，如果我只关注人脸区域，然后每一帧都做人脸的数据包匹配应该计算量很大很复杂吧！ 至于您说的目标检测器应该是针对帧图像来做的，而我想要一帧图像中的特定区域，我需要识别出特定区域的在该帧中代表数据包，不知道是否有更成熟的方案或者工具可以实现？

不完全是这样，ROI区域要100%传输，非ROI区域并不是不传输，而是选择性传输，也就是说当我检测到该数据包属于ROI区域时，我必须传输！但是非ROI区域则会选择性丢弃一些！所以不能直接把ROI提取出来生成新的文件传输，那样非ROI区域的数据就都丢失了！

> I think this is a common use case, not only in your company's case. Server side version and client side version may not upgrade synchronized in real world, server...

> Hower the client side may be a little difficult, and even for me it’s a bit vague to what extent the function needs to be implemented, the following pdf...

If you still want to handle the replica version, you can let meta to sync the replica server in `config_sync` and return the client. In short, I think it will...

## 1 > The java client may pull outdated feature list data, so we need a timer repeated scheduled function process used to continuously update the local feature list periodically......

> meta server should remember the feature list of all the replica server Well, I see, you can select to resolve it based different cases. Another resolution is the response...

> java client will not get the no support response Yeah, I see, you are right

> > > java client will not get the no support response > > > > > > Yeah, I see, you are right > > so sadly, I should...

In today's(2021.3.2) test, the bug recurred