使用本地镜像文件进行Deeplab测试

Question

我正尝试在本地计算机上运行deeplab。我在我的电脑上安装了deeplab并定义了路径。Deeplab在我的本地计算机上运行成功。这是deeplab_demo.ipynb的最后一块代码

SAMPLE_IMAGE = 'image1'  # @param ['image1', 'image2', 'image3']
IMAGE_URL = ''  #@param {type:"string"}

_SAMPLE_URL = ('https://github.com/tensorflow/models/blob/master/research/'
               'deeplab/g3doc/img/%s.jpg?raw=true')


def run_visualization(url):
  """Inferences DeepLab model and visualizes result."""
  try:
    f = urllib.request.urlopen(url)
    jpeg_str = f.read()
    original_im = Image.open(BytesIO(jpeg_str))
  except IOError:
    print('Cannot retrieve image. Please check url: ' + url)
    return

  print('running deeplab on image %s...' % url)
  resized_im, seg_map = MODEL.run(original_im)

  vis_segmentation(resized_im, seg_map)


image_url = IMAGE_URL or _SAMPLE_URL % SAMPLE_IMAGE
run_visualization(image_url)

我正在运行jupyter笔记本。它适用于'_SAMPLE_URL‘变量或任何包含图像文件的链接。我想用一个本地镜像文件来测试deeplab。

注意:已经定义了环境变量。这样我就可以从这个笔记本访问我的本地文件了。所有的库都工作正常。但我不想在URL上测试图像，只想测试本地图像文件。

Answer 1

更改本地路径

_SAMPLE_URL = ('https://github.com/tensorflow/models/blob/master/research/‘'deeplab/g3doc/img/%s.jpg?raw=true')

->

_SAMPLE_URL = ('file:///path to/tensorflow/models/blob/_SAMPLE_URL/_SAMPLE_URL/‘'deeplab/g3doc/img/%s.jpg')

Answer 2

我通过在'original_im = Image.open(BytesIO(jpeg_str))‘行将'BytesIO(jpeg_str)’改为'123.jpg‘来完成。