UiPath Activities Guide

Computer Vision Recorder

Once you install the Computer Vision activities pack, the Computer Vision Recorder wizard becomes available in the Ribbon. This recorder is suitable for automatically generating workflows that use the CV activities, offering you the full spectrum of capabilities this package has to offer. To find out more what situations you can use the Computer Vision activities for, go here. Once opened, the recorder looks like this:

The Settings button enables you to configure the connection details of the CV server as well as the default OCR engine that you want to use.

By default, the recorder is set to use the https://cv.uipath.com community server. To connect to an on-prem deployment, you must change the URL in the Server field to your server's URL. Connecting through the UiPath Cloud service requires you to add your key in the ApiKey field. For more details, visit the Deploying a Local Machine Learning Model page.


The settings regarding the server connection are project-wide, and are reflected in all subsequent CV Screen Scope activities.

Clicking the Record button enables you to indicate the application you want to automate, which is used as a target for all subsequent CV activities. Once an application is indicated, the selected window is sent to the CV server, where it is processed, and the wizard enters recording mode. All other screen content besides the selected window is grayed out, and the recorder interface changes:

The Type button enables you to select an identified text field and type text and/or hotkeys into it. After selecting a field, a window enabling you to customize the type action is opened.

You can specify the text you want to type in the text box, as well as add in hotkeys by using the + button. The Empty field check box deletes all the text inside a field before typing into it. Once finished, pressing Enter saves the action.

The Click relative button enables you to click any UI element. If necessary, the wizard can also set an anchor for the area you want to click.

The Get button extracts the text from a specified area. This action can also request an anchor if necessary.

The Save & Exit button closes the recording session and generates the activities required to perform the actions you recorded in the Designer panel.

The Stop Recording button stops recording actions, takes the target window out of focus, but does not close the wizard. This is useful at design-time if you want to perform changes to the application you are automating. Once clicked, the wizard looks like this:

To continue recording in this state, you must click the Record button again.

The Refresh button can be used when the target window changes and needs to be resent to the CV server. This can happen, for example, after clicking a button that enters another section of the application.

The Change Application button enables you to indicate another application window, in case you want to switch to automating another application.

Computer Vision Recorder

Suggested Edits are limited on API Reference Pages

You can only suggest edits to Markdown body content, but not to the API spec.