Memory Video Generation Using Character Feature Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating picture collections or videos based on themes in intelligent electronic devices fail to consider user preferences, leading to a poor user experience as users cannot view pictures or videos of characters they like.
Innovation Solution
A method for generating memory videos that allows users to actively select characters of interest through feature extraction and operation on a memory entry area, enabling the electronic device to search and generate videos tailored to the user's preferences, thereby improving user experience.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If the system automatically generates picture collections or videos based on themes without user input, then the generation process is simple and fast, but the user cannot view pictures or videos of characters they like, resulting in poor user experience
Solution Approach 1:
The system performs feature extraction on characters in pictures in advance and stores character features before video generation. When a user requests a memory video, the system can quickly search for pictures containing specific characters using the pre-extracted features, avoiding time-consuming feature extraction during video generation while still enabling user-specific character selection
Solution Approach 2:
The patent introduces a character feature database as an intermediary between user input and video generation. The system extracts character features from pictures and stores them in this database, then uses the database to quickly retrieve relevant pictures when generating memory videos for specific characters, thus enabling personalized video generation without significantly increasing generation time
2Ease of operation
If the system requires users to manually select and tap multiple pictures to create a collection, then the video can be customized to user preferences, but the number of user operations increases significantly
Solution Approach 1:
The patent replaces the mechanical manual selection process with an automated intelligent search system. Instead of requiring users to manually tap and select multiple pictures, the system uses character feature extraction and matching to automatically search for and select relevant pictures, substituting user manual operations with automated computer vision and pattern matching algorithms
Solution Approach 2:
The system performs self-service by automatically searching for and selecting pictures that contain the user's selected character. After the user selects a character from the first picture, the system autonomously searches the picture library, identifies pictures containing the same character through feature matching, and compiles them into a memory video without requiring further user intervention
3Measurement precision
If the system browses pictures individually without video playback, then each picture can be viewed in detail, but the user experience is fragmented and memory connection is lost
Solution Approach 1:
The patent merges multiple individual pictures into a unified video format. Instead of presenting pictures as separate, disconnected items that require individual viewing, the system combines selected pictures into a continuous memory video that maintains both the detail of individual pictures and the continuity of memory through sequential playback with transitions
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
This application relates to the field of picture processing, and provides a method and an apparatus for generating a memory video, and an electronic device. Through a first operation of a user on a first picture, the electronic device displays a second display interface including a memory entry area, and through a second operation of the user on the memory entry area, the electronic device searches for pictures matching at least one character of the first picture, to finally generate a first memory video dedicated to the at least one character. In this way, through the first display interface and the first operation, a condition of actively triggering a memory function is friendly provided for the user, and through the second display interface and the second operation, a condition of generating the memory function is friendly provided for the user, so that the user can perform, based on a requirement of the user, an operation on pictures of characters that the user is interested in, to generate the memory video, so as to meet an actual requirement of the user, thereby better improving user experience.