GENERAL AI

GENERAL AI can proceed with auto-labeling immediately without manual labeling by utilizing AI Model that has already been developed.

Start Auto-Labeling with GENERAL AI

1. Start Auto Labeling

1. Click the Dataset in the top navigation bar.

2. Select the labeling project that you created in the previous step.

3. On the Dashboard tab of the Labeling project, click the Start Auto Labeling button.

4. Sets options for GENERAL AI auto-labeling.

5. After setting the options, click the Start Auto Labeling button to start auto labeling automatically.

6. You can check the progress of auto-labeling on the dashboard, and we will notify you by e-mail and notification when it is completed.

GENERAL AI Auto Labeling Options

Select the type of AI

Select labelling AI that you want to develop from General AI, including human, Animal, Self-Driving, FacePoint Detection, and Human Keypoint.

Select the Class

You can select the object that you want to recognize among the objects in each category. At least one class must be selected and multiple selections are possible.

Select Auto Labeling Type

• Bounding Box : Labels objects in a box shape of a rectangle.

• Polygon: Mark object attention as a dot according to the shape of the object and label it linearly.

• Semantic Segmentation : Labels objects by pixel.

The Number of Auto Labeling Image

Auto-labeling is possible for the total number of images uploaded from a minimum of 100 pages.

2. Check Auto Labeling Results

1. In the Labeling Project's Data List tab, set the filter at the bottom right of the tab menu to Complete Task Status.

2. Click the data you want to view in the Data List to see the auto-labeling results, and click the Inspect Label button in the bottom right to modify the labeling. (You can find out how to use the labeling tool in Labeling > Manual-Labeling > Object Detection.)

Auto-Labeling Types supported by GENERAL AI

Recognition of human

Detects the location of the human object in the image and provides information.

• Identification Class Name : person

• Auto-Labeling Format : Bounding Box, Polygon

Recognition of Animal

Detects the location of the animal object in the image and provides information.

• Identification Class Name : person, bear, bird, cat, cow, dog, elephant, giraffe, horse, sheep, zebra

• Auto-Labeling Format : Bounding Box, Polygon

Autonomous Driving Semantic Segmentation

It detects specific objects in the transportation and road facilities in the image, identifies locations, and provides information.

• Identification Class Name : person, airplane, bench, bicycle, boat, bus, car, firehydrant, motorcycle, parkingmeter, stopsign, trafficlight, train, truck

• Auto-Labeling Format : Bounding Box, Polygon, Semantic Segmentation

Facepoint detection

The image recognizes features (landmarks) such as nose, eyes, mouth, eyebrows, and jawline in the form of points.

• Identification Class Name : person

• Auto-Labeling Format : Bounding Box, Polygon

Human Key Points Detection

It detects key points by placing coordinates on the human joint in the image and connecting them with lines. This allows you to classify people's poses, etc.

• Identification Class Name : person

• Auto-Labeling Format : Bounding Box, Polygon