Foreground Object Outline Detection in Image Processing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current image processing technologies for online applications, such as shopping and live streaming, limit user interaction with displayed images and videos, offering only minimal forms of engagement.
Innovation Solution
An image processing method that uses a machine-learning model, specifically a semantic segmentation model combining a Mobile GPU-Aware (MoGa) network and a Lite reduce Atrous Spatial Pyramid Pooling (LR-ASPP) segmentation head, to generate a mask image, differentiate between foreground and background objects, and enhance image interaction by determining an outline for user interaction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional image display methods are used in online applications, then the system is simple and easy to implement, but user interaction with images is limited and engagement is low
Solution Approach 1:
The image is segmented into foreground objects and background using semantic segmentation. A mask image is generated where pixels corresponding to foreground objects are set to a first grayscale value range and background pixels to a second grayscale value range, enabling differentiated interaction with image components
Solution Approach 2:
A machine learning model (semantic segmentation model) is introduced as an intermediary to process the input image and generate the mask image. This intermediary enables the system to understand and differentiate image content, facilitating enhanced user interaction without directly modifying the display interface
2Adaptability or versatility
If semantic segmentation models are used to generate mask images and enable interactive modes, then user engagement and interaction modes are enhanced, but computational resources and processing time increase
Solution Approach 1:
The system applies semantic segmentation selectively - generating mask images and enabling interactive features only when foreground objects are detected in the input image. This partial application reduces unnecessary computational overhead while maintaining enhanced interaction capabilities when needed
Solution Approach 2:
The system changes the parameter representation of image data by converting color images to grayscale mask images with specific value ranges. This transformation simplifies subsequent processing and enables efficient differentiation between foreground and background regions
Data Source
AI summary
Embodiments of this specification provide methods and apparatuses for processing images, devices, and storage media. A method includes: obtaining an image for processing; generating a first mask image corresponding to the image based on a machine-learning model and determining whether the image has a foreground object; in response to determining that the image has the foreground object, setting pixels corresponding to the foreground object to a first grayscale value range, and setting pixels corresponding to non-foreground objects in the first mask image to a second grayscale value range; determining an outline corresponding to the foreground object in the image according to a grayscale value range difference between the first grayscale value range of the foreground object and the second grayscale value range of the non-foreground objects in the first mask image; and sending, to a client device for display, the image with the determined outline for interaction by a user.