hello, I used your model to infer a picture from the internet, however, low resolution picture seems to work bad, but high resolution picture works well, and I read your paper, you paper say the network works well for various poses and scales of human in the foreground, do you know what is the reason of bad results in low resolution images?
hello, I used your model to infer a picture from the internet, however, low resolution picture seems to work bad, but high resolution picture works well, and I read your paper, you paper say the network works well for various poses and scales of human in the foreground, do you know what is the reason of bad results in low resolution images?