(Old)Annotation Templates

Intoduction

This page introduces a list of templates provided by the ABEJA Platform Annotation tool.

More templates will be added by Annotation Team based on your request. Request from chatbox. Reference is here for sample of various output results

1. Image Classification
2. Object Detection
3. Text Classification
4. Video Scene Labeling (Video)
5. Image Segmentation
6. Voice-to-Text
7. Image-to-Text
8. Polygon Segmentation
9. Text Extraction
10. Polygon Segmentation_selectable
11. Video Detection
12. Audio Region Labeling(Coming Soon)

1. Image Classification

• Classify image on this template.
• Assign preset tag for image.
• Ex.: Product classification, Emotion classification and Assignment of attribute information
• Supported format: jpg,png
• Output results: json <sample>

2. Object Detection

• Detect objects in an image on this template.
• Draw bounding box on an image
• Ex.: Product detection, Object counting and Face detection
• Supported format: jpg,png
• Output results: json <sample>

3. Text Classification

• Classify text body on this template.
• Add reset tag to text body.
• Ex.: Text category classification / text harmful classification / text emotion classification
• Work image movie
• Supported format: txt, json ※Less than 1,500 Characters(Japanese) for 1 file is recommended.
• Output results: json <sample>

4. Video Scene Labeling

• Classify scene in the video on this template.
• Play video to get start / end sequence information
  Add that sequence information and preset tag information to extract scene information
• Ex.: Video element extraction and cyclic analysis
• Supported format: mp4
• Output results: json <sample>

5. Image Segmentation

• Extract a regions in an image on this template.
• Paint regions you want to extract in an image.
• Extract an object shape in the picture.
• Draw a polygon while hitting the apex of the object you want to detect in an image, and paint that area.
• Ex.: Region extraction, Area extraction and Damaged extraction
• Supported format: jpg,png
• Output results: json,png <sample>

6. Voice-to-Text

• Extract text data from speech on this template.
• Play audio and add information in free text on it.
• Ex.: Text extraction, speech category classification, speech emotion classification, native check
• Supported format: mp3
• Output results: json <sample>

7. Image-to-Text

• Extract text information from an image on this template.
• Information will be added to an image as a free text.
• Ex.: Optical Character Recognition(ORC)/ Attaching Metadata to images
• Supported format: jpg.png
• Output results: json <sample>

8. Polygon Segmentation

• Extract object shape in the image on this template.
• Extract coordinate information by drawing a polygon while pointing the vertices for the object you want to detect in an image.
  Polygon Segmentation is able to get a shape information or create parallelograms and trapezoids witch Object detection is not able to do.
• Able to select *¹ Flat mode or *² Line mode
  *¹ Create the polygon that connect start to end　² Create the line non connected start to end
• Polygon Segmentation is able to set any label on work screen.
• Ex.: Object frame extraction, character frame extraction
• Supported format: jpg,png
• Output results: <sample>

9. Text Extraction

• Add a preset tag to the word you want to extract in the text to extract words and classify words on this template.
• Ex.: Word extraction, word classification
• Work image movie
• Supported format: txt
• Output results: json <sample>

10. Polygon Segmentation_selectable

• Extract object shape in the image on this template.
• Extract coordinate information by drawing a polygon while pointing the vertices for the object you want to detect in an image.
• *¹ Flat mode or *² Line mode
  *¹ Create the polygon that connect start to end　² Create the line non connected start to end
• Polygon Segmentation_selectable is able to select a redefined label.
• Ex.: Object frame extraction, character frame extraction
• Supported format: jpg,png
• Output results: json <sample>

11. Video Detection

• Divide the video into image in frame and detect objects
• Draw unique bounding box for consecutive images
• Ex.: Detect tracking objects in video / Detect number of unique object
• Supported format: mp4
• Output results: Json <sample>

12. Audio Region Labeling（Coming soon）

• Coming Soon