Tutorial

Installation

To start using PCloudCV is easy. You can download PCloudCV using:
https://github.com/batra-mlp-lab/pcloudcv/wiki/Setup

You need to make sure you have redis-server installed and running before you can use the python APIs.

Please follow this link to install it: http://redis.io/topics/quickstart

Sample Application:

Steps to write a simple application using CloudCV APIs can be seen in run.py

1. Import PCloudCV class from pcloudcv module

       from pcloudcv import PCloudCV

2. Start by creating a dictionary containing three parameters:

   1. Full Path to Input Folder where all the images are present, ignore it if you want to use path mentioned in config file. Prefix the path with "local:" if its a local path or "dropbox:" if its a dropbox path
   2. Full Path to Output Folder where the results will be stored, ignore it if you want to use path mentioned in config file.
   3. Algorithm that needs to be run. More information about the valid names are [here] 1. Ignore it if you want to use the name of the executable mentioned in the config file

   config_dict =    {'input':'local:/path/to/input/folder','output':'/path/to/output/folder','exec':'executable_name'}

3. Create a variable to store the full path to the config file. More information about config files [here] 2. Also create a variable to mention whether you want to use APIs after logging in or not.

      path_to_config_file = '/home/dexter/projects/vt/pcloudcv/pcloudcv/config.json'
      login_required = True

4. Create an object of the PCloudCV class

      pcv = PCloudCV(path_to_config_file, config_dict, login_required)

5. Start uploading images in the background using:

      pcv.start()

6. The APIs will upload the result in the background. It will wait for results in the background so that your main application thread is not stuck. When the results are available, they will automatically be retrieved from the server and the resultant text file/image file will be downloaded.

---

Complete Code

from pcloudcv import PCloudCV
config_dict = {'input':'/path/to/input/folder','output':'/path/to/output/folder','exec':'executable_name'}
path_to_config_file = '/home/dexter/projects/vt/pcloudcv/pcloudcv/config.json'
login_required = True
pcv = PCloudCV(path_to_config_file, config_dict, login_required)
pcv.start()