Configuration

In the Configuration tab, you have the ability to modify various settings.

The project name, which is entered when creating the project, is displayed on this page. The name can be changed in the configuration section. Additionally, tags can be added to the project in the designated field, which may contain additional information about the project. Memo can also be entered in the notes section.

You can specify the minimum object size, if necessary.

Moreover, the automatic assignment of workers can be selected, which allows for the same worker to be assigned to all stages of file annotation and verification. Alternatively, the "Assign different verifiers to each step" option can be chosen to assign different users to verify each step. This is useful when double verification is necessary, and each file needs to be verified by multiple reviewers from the same pool.

The briefing page address and instruction link are also provided on this page. Additionally, special permissions can be granted to clients and annotators. Clients can be permitted to edit annotations in the project, while annotators can be allowed to see and modify external comments. These permissions are not active by default, however. Finally, the validation script can be entered in the designated field.

You can choose ML Models here as well.

Layers in image editing are individual components that can be stacked on top of each other to create complex images. Each layer can contain different elements like text, images, graphics, effects, and more. You can edit and modify each component of the image independently using the layer control.

A layer is a separate level of an image. When working with image settings, layers are added based on the image's configuration.

URL configuration for Layer image is similar to the configuration of Frame image URL. You only need to use the variable e.g.{layerid} that will lead the platform to the correspondent file of the layer.

In the editor, you can control each layer by scrolling and adjusting its transparency.

This feature is used for various types of images across different fields, especially when you need to combine and manipulate multiple elements to achieve a specific visual effect or to analyze data.

It is possible to specify the minimum width and height of the objects, if necessary.

Moreover, the automatic assignment of workers can be selected, which allows for the same worker to be assigned to all stages of file annotation and verification. Alternatively, the "Assign different verifiers to each step" option can be chosen to assign different users to verify each step. This is useful when double verification is necessary, and each file needs to be verified by multiple reviewers from the same pool.

The briefing page address and instruction link are also provided on this page. Additionally, special permissions can be granted to clients and annotators. Clients can be permitted to edit annotations in the project, while annotators can be allowed to see and modify external comments. These permissions are not active by default, however. Finally, the validation script can be entered in the designated field.

The platform provides the ability to enter multi-line text in text attributes. To enable this, simply check the checkbox next to the corresponding “Editor behavior” option on the project configuration page.

This improves the readability of long values and simplifies working with large amounts of text.

The OCR (Optical Character Recognition) tool enables automatic text extraction from images. When this option is active, users can:

Once processed, the recognized text is automatically populated into the corresponding attributes.

The platform features an integrated translation tool powered by Gemini models, supporting seamless translation between Ukrainian, English, and Russian with automatic language detection.

Two processing modes are available:

The File Mapping feature allows users to change default platform filenames to custom ones during export using a CSV mapping file.

This provides full control over the exported dataset's structure without altering the original project data, helping to ensure data anonymization and maintain proper frame sequencing in case of incorrect naming.

Also, using the Download button, users can download a previously uploaded CSV file for further editing and re-uploading to the platform.