IVPD supports the following features.
IVP includes the following features:
Image editing
Application scenario | Feature description | Reference |
Image matting/segmentation | Segments the main entity in an input image and outputs the result as a transparent PNG file. It supports entities such as humans, animals, and objects. |
|
Image super-resolution/sharpening | Enlarges an input image by four times while maintaining the definition of the resulting image based on inferred details. |
|
Image cropping | Transforms an input image to a specified size. It automatically detects the main entity's region and applies the optimal crop. |
|
Image color transformation | Transforms the colors of an input image automatically or based on a specified color palette. It also prevents abnormal color spreading in visual hot spot areas. |
|
Image style transfer | Transfers the style from a specified style image to an input image. This changes visual styles such as the color and brushstrokes of the input image. |
|
+
=
Image analysis
Application scenario | Feature description | Reference |
Image color palette analysis | Analyzes the color information of an input image. It provides the color values in RGB and HEX formats and their corresponding percentages. | Input image: |
Image style analysis | Analyzes the style type of an input image and provides possible style tags, such as visual style: Chinese style, and semantic style: elegant. | Input image: |
Image element detection | Detects elements in an input image, annotates their positions with bounding boxes, and classifies them by their primitive data type, such as person/object, decoration, or text. |
|
Extracted color card: 
Detected style tags: Style type: Visual style: Chinese style, Semantic style: elegant