Video Call Visual Object Recommendation System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Video call services often experience awkwardness due to unfamiliarity between participants, leading to tense conversations, especially in random matching scenarios, where traditional icebreakers are insufficient in alleviating this tension.
Innovation Solution
A method and system that utilize machine learning to recognize relationships among characters, backgrounds, and visual objects during video calls, automatically recommending appropriate visual objects based on appearance, background, time, location, profile, and voice-related conditions to enhance the conversation atmosphere.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional icebreakers are used in video call services, then conversation atmosphere can be improved, but the effectiveness is insufficient especially in random matching scenarios
Solution Approach 1:
The system dynamically selects visual objects based on real-time analysis of video call conditions including character appearance, background, time, location, profile information, and voice characteristics. This dynamic adaptation allows the icebreaker to effectively respond to varying situations rather than using static predetermined content.
Solution Approach 2:
The patent replaces traditional manual icebreaker selection with an automated machine learning system that analyzes multiple parameters simultaneously. This substitution of automated intelligent systems for manual or traditional methods enables more reliable and effective tension alleviation through comprehensive situation assessment.
2Ease of operation
If automatic visual object recommendation system is implemented, then conversation atmosphere is enhanced, but system complexity increases
Solution Approach 1:
The system segments the complex task of icebreaker selection into multiple independent analysis modules: character appearance analysis, background analysis, time analysis, location analysis, profile analysis, and voice analysis. Each module processes specific parameters independently, then their results are integrated to determine the final visual object recommendation.
Solution Approach 2:
The patent introduces a machine learning model as an intermediary that processes multiple input parameters and translates them into visual object recommendations. This intermediary layer manages the complexity by providing a standardized interface between diverse input data sources and the output recommendation system.
3Measurement precision
If multiple parameters are analyzed for visual object selection, then recommendation accuracy is improved, but processing time increases
Solution Approach 1:
The system performs preliminary processing of input parameters such as extracting character features from video frames, analyzing background images, and preprocessing voice signals before the actual visual object selection. This preliminary action prepares data in advance, reducing the computational burden during real-time recommendation and minimizing processing delays.
Data Source
AI summary
Disclosed are a method of learning relationship among characters in video call or its background, temporal and spatial information, and visual objects and automatically recommending and providing a visual object using the relationship, and a system configured to execute the method. A method of providing video call may include: storing a visual object selection model including relation information between at least one visual object and at least one selection factor, by a video call providing system; and automatically determining, by the video call providing system, a recommended visual object to be displayed on at least one of a terminal and a counterpart terminal of the terminal performing video call, at a point of time specified for displaying a visual object, based on the visual object selection model.