Javacr
Javacr
> 此外，我之前有做过细粒度图像分类，其中采用的attention机制以及实验设定供参考：https://arxiv.org/abs/1901.00392 大佬涉及方方面面，佩服。ghostnet的确好用，快而准。同时，谢谢大佬提供的论文链接，这两天我研究研究。
> > (base) kakutooru@MacBook-Air PyQt5-YOLOv5 % pip install PyQt5-tools > > Looking in indexes: https://pypi.douban.com/simple/ > > Collecting PyQt5-tools > > Downloading https://pypi.doubanio.com/packages/b7/70/a25aab849a5e38ae78716acde1f917a14d9d819c8600bbb8cf536781f7c5/pyqt5_tools-5.15.4.3.2-py3-none-any.whl (29 kB) > > Using cached https://pypi.doubanio.com/packages/36/b8/a255b8160e863678aa7d559a4c8c33b5448a6e270992d1464ec9f87ed633/pyqt5_tools-5.15.4.3.1-py3-none-any.whl...
> This might be caused by the image quality. I would strongly recommend you use RGB as input. The better the image quality, the better the performance (You can try...
Here is result video: https://drive.google.com/file/d/1Ba7jmtGqcdpPgYocD49N3mlw2qo2gdMc/view?usp=sharing
    red represents left hand, blue represents right hand
    In fact, most of the time, the results are correct. The bad results are due to the limitations of the algorithm itself.
> Hi @Javacr, > > Thank you for bringing this issue to our attention. We are aware of the problem, as noted in [this GitHub issue](https://github.com/google-ai-edge/mediapipe/issues/5156). For now, we are...
> > 同，请问您解决了吗 > > 没，在csdn上有人说改dataset，但是不知道怎么弄 你重新拉一下仓库，已经更新了
> 不知道为什么问题描述部分的一些文字会出现删除线 这个问题，我前几天也遇到了，原因是版本问题。我将分支切换到r1.4，paddlepaddle-gpu用2.4.0，成功解决该问题。