33 STEPS
1. The first step is to open Cogniac homepage and select Detection from the Vision AI Applications. You may notice many different options but for now we will select Detection under Vision AI Applications.
2. Select the Name text field to create a name of the Detection Application.
3. Click Next button, and you will proceed to the Input Subjects page.
4. An Input Subject is simply a group or collection of images or videos that will be afterward used by the Application for Training. The Input subject is the media that has not yet been processed by the Detection Application.
5. Type an Input Subjects name. Again, please consider a name that will help you differentiate the Input Subject in the future.
6. Select the Create a new subject which in the current scenario is called 'animals'.
7. Click the Next button, so you can proceed to the Output Subjects page.
8. The Output Subject of the Detection application is where you will find the labeled and sorted images. Usually, the Output Subject gets populated with media (images or videos) after successful Training of the Application.
9. Type Output Subjects name. Again please consider a name, that will be directly related to the Subject Of Interest displayed within the image.
10. Click Create a new subject to create the first Output Subject, that in the current case is named 'cat'
11. Type the second Output Subjects name.
12. Select Create a new subject to create the second Output Subject. In the current case, it is named 'dog'
13. Click Next
14. Click Frames Per Second Requirement to change the value.
15. Type the Frames Per Second Requirement in a numeric format.
16. Scroll down and click Complete
17. Select the Input Subject, which is always displayed in the beginning of the Application Pipeline. In current scenario select 'animals'.
18. Click Upload Media to upload the Images that will be used to train the Application Model.
19. Click select files or select folder to upload the initial Dataset of images/ videos to the Input Subjects of the Application
20. Click Next once you have selected the files or the folder containing the Media
21. Click Upload
22. Click on the Application Name in our case it is animals-detector
23. Click Replays, so you can provide feedback, since the Training is a continuous process of uploading images and Providing Feedback.
24. Select the input subject which in the current scenario is 'animals'.
25. With your mouse drag the Probability Filter, and drop it at the beginning of the scale if you need to force the Feedback for the images with 0% to 100% probability.
25b. Drop
26. Scroll down and select the Force Feedback option
27. Click Replay Subjects once everything is completed.
28. Select the Application name which is in the middle of the Application Pipeline. In current case it is 'animals-detector'
29. Select the Provide Feedback, button. We can see that in the Feedback queue there are 5 elements (media) pending feedback.
30. Select the Subjects Of Interest displayed within the image. In the current step select cat.
31. Then select also 'dog', since a dog is also there displayed in the image.
32. Once you've selected all the Subjects Of Interest (please keep in mind, they could be more than two), select the Confirm blue button in the end.
33. During providing feedback there are also: Back, Sideline and Skip options.
