When you draw a box around text in a video, the video Optical Character Recognition (OCR) template automatically recognizes the text inside the box.
Data format examples
In CSV and XLSX files, each column represents a dataset field. In Manifest files, each subfield under the data field represents a dataset field. You can use custom Field Names. When you configure the dataset, you must select the corresponding field names.
CSV and XLSX formats
video_url |
oss://****.oss-cn-hangzhou.aliyuncs.com/iTAG/mov/1.mp4 |
oss://****.oss-cn-hangzhou.aliyuncs.com/iTAG/mov/2.mp4 |
For a demo file, see video_data_demo.csv.
Manifest format (JSONL format)
{"data":{"video_url":"oss://****.oss-cn-hangzhou.aliyuncs.com/iTAG/mov/1.mp4"}}
{"data":{"video_url":"oss://****.oss-cn-hangzhou.aliyuncs.com/iTAG/mov/10.mp4"}}For a demo file, see video_data_demo.manifest.
Limits
Supported video formats: MP4, WebM, and OGG.
The recommended video length is 5 minutes or less.
Configuration
Annotation area (Required)
You can use Video to Annotate to configure the video field.
Click Video to Annotate and select a dataset. Then, configure the parameters in the Configuration area on the right. The following table describes the parameters.
Parameter |
Description |
Title |
Do not modify the default title area. |
Dataset Field |
Select the video URL field from the dataset fields. You can customize the field name. |
Prompt |
Do not modify the default prompt area. |
Card Style Configuration |
This setting is not configurable and has no effect. |
Global configuration
You can use Global Configuration to configure the User-Defined Functions (UDFs) for the template. For example, you can configure a UDF to retrieve the video frames per second (fps).
Click the settings icon 
to configure the parameters on the right. The following table describes the parameters.
Parameter |
Description |
Title |
Do not modify the default title area. |
Component Version |
This setting is not configurable and has no effect. |
Pre-annotation |
Configure the UDF to get the video fps in this section. The configuration must be the same as that shown in the figure. |
Data Validation |
This setting is not configurable and has no effect. |
Other
Click Back to return to the template selection page.
Click Intelligent Configuration to configure services such as pre-annotation and UDFs.
Click Discard Configuration to specify whether samples can be discarded.
Click Import Dataset to import a different dataset.
Click Export Template to export the template in JSON format.
Click Save Template to save the current template configuration.
Click Create Task to save the current template configuration and create a task.
Operation demo
Draw a box around the text in the video and double-click the box. A pop-up window appears with the text automatically recognized by OCR. Create a new ID and save it. The data is then displayed in the video, the Q&A area on the right, and the timeline at the bottom.
Find the keyframes where the content for an ID appears. Draw a box, select the same ID, and save. The data in all three areas updates.
Double-click a box in the video to edit or delete the content for the corresponding ID. In the Q&A area on the right, hover over the
icon in the operation column to view the keyframes for the ID. Click a keyframe to jump to it. Click the 
icon in the operation column to delete the data for the ID.Hover over a data bar on the timeline to view its full text. Double-click a data bar to jump to its starting keyframe.