1. Quick start with example dataset ¶

Let’s use ReNomIMG with an example dataset.

We will use the PASCAL VOC dataset in this example.

           
            The PASCAL Visual Object Classes (VOC) Challenge
           
           Everingham, M., Van Gool, L., Williams, C. K. I., Winn, J. and Zisserman, A.
          
           International Journal of Computer Vision, 88(2), 303-338, 2010
          
           PDF:
           
            http://host.robots.ox.ac.uk/pascal/VOC/pubs/everingham10.pdf
           
           Web site:
           
            http://host.robots.ox.ac.uk/pascal/VOC/

1.1. Prepare Dataset ¶

First we need to prepare the image dataset. Here we use the Pascal VOC dataset. You can download and prepare the dataset by running the following command.

            # This command will download the data and save it to the current directory.
renom_img setup_example

# ### Setup a example dataset ###
# 1/7: Downloading voc dataset.
# 2/7: Extracting tar file.
# 3/7: Moving image data to datasrc/img...
# 4/7: Moving image data to datasrc/prediction_set/img...
# 5/7: Moving xml data to datasrc/label/detection...
# 6/7: Moving segmentation target data to datasrc/label/segmentation...
# 7/7: Creating classification target data in datasrc/label/classification...
# Setup done.

           

The datasrc and storage directories will be created in the current directory.

1.2. Run ReNomIMG ¶

Once the dataset is ready, start the ReNomIMG server. The command below runs ReNomIMG on port 8080.

renom_img --port 8080

Next, let’s move on to the section “How to use the ReNomIMG GUI”.