Instant Messaging Expression Image Retrieval via Object Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing instant messaging applications face inefficiencies in saving and displaying expression images, particularly when users have a large quantity of favorited images, making it time-consuming to find appropriate images for current chat situations.
Innovation Solution
An image data processing method and apparatus that, in response to an expression image trigger operation, extracts target object features from image data, retrieves a set of associated expression images from an object feature library, and displays these images, allowing users to select and send the appropriate expression image.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If users save a large quantity of expression images, then the variety of expression images increases, but the time required to search for appropriate images increases
Solution Approach 1:
The system performs preliminary classification of expression images by extracting object features and organizing them into categories before the user needs to search. When a user triggers an expression image search, the system quickly retrieves pre-classified images based on the trigger image's object features, avoiding the need to search through all saved images from scratch.
Solution Approach 2:
The patent introduces an object feature library as an intermediary between the user's expression image collection and the search function. The library stores extracted object features and enables efficient matching between trigger images and candidate expression images, significantly reducing search time while maintaining access to a large quantity of images.
2Ease of operation
If users manually search through all favorited expression images, then the user can find appropriate images, but the operation complexity and time consumption increase
Solution Approach 1:
The system performs automatic object feature extraction and image matching without requiring manual user intervention. When a user triggers the expression image function, the system automatically extracts features from the trigger image, queries the object feature library, and retrieves matching expression images, making the process self-service and highly efficient.
Solution Approach 2:
The patent replaces the mechanical manual searching process with an automated computer vision system that uses object detection and feature matching algorithms. This substitution transforms the manual browsing and selection process into an automated retrieval process based on object features, dramatically reducing time consumption.
3Adaptability or versatility
If the system displays all favorited expression images, then the user has access to all options, but the display efficiency and user experience deteriorate
Solution Approach 1:
The system extracts and displays only the relevant expression images that match the object features of the trigger image, rather than displaying all favorited images. This extraction approach maintains adaptability by providing context-appropriate options while significantly improving productivity by reducing the number of images the user needs to review.
Solution Approach 2:
The patent applies local quality by providing different image sets based on the specific context - when a trigger image is provided, the system displays expression images with matching object features; without a trigger image, it may display recently used or popular expressions. This contextual adaptation maintains versatility while optimizing retrieval efficiency for each specific situation.
Data Source
AI summary
Embodiments of the disclosure provide an image data processing method and apparatus, an electronic device, and a storage medium. The method includes: obtaining, in response to an expression image trigger operation of a target user, an operation object associated with the expression image trigger operation on a session page of an instant messaging application; based on the operation object being image data, extracting a target object feature in the image data, and obtaining a first image set associated with the target object feature from an object feature library; and determining a target expression image in response to a selection trigger operation of the target user in the first image set, and displaying the target expression image on the session page.