ML-Based Gameplay Highlight Selection for Continuous Recording
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current video game recording methods require user intervention to capture interesting moments, breaking immersion and relying on manual selection, as they typically cyclically overwrite recordings, storing only the last few minutes of gameplay.
Innovation Solution
A data processing apparatus and method using machine learning models to automatically select interesting portions of gameplay by analyzing image and audio data, generating a recording comprising selected groups of image frames based on interest scores from multiple models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If continuous recording is stored in temporary buffer with cyclic overwriting, then storage space is conserved, but interesting moments may be lost due to overwriting
Solution Approach 1:
The system performs preliminary analysis of recording content using machine learning models to identify interesting moments before final storage decisions are made. This allows the system to proactively detect and preserve important segments before they would be overwritten, resolving the contradiction between limited storage and reliable capture of interesting moments
Solution Approach 2:
The system implements feedback mechanisms where machine learning models continuously analyze the recording stream and provide real-time identification of interesting moments. This feedback loop enables dynamic adjustment of storage priorities, ensuring that identified interesting segments are preserved while managing limited storage resources efficiently
2Manufacturing precision
If manual selection of recording portions is required, then recording quality is improved, but user immersion is broken and operation complexity increases
Solution Approach 1:
The system performs self-service by automatically analyzing and selecting interesting recording portions using machine learning models without requiring user intervention. The models independently evaluate content based on predefined criteria and generate highlight recordings autonomously, maintaining both high quality and user immersion simultaneously
Solution Approach 2:
The system replaces the mechanical interaction of manual button pressing with automated machine learning-based selection. The ML models substitute for user action in identifying interesting moments, eliminating the need for users to break immersion and manually select segments while preserving recording quality through intelligent automated analysis
3Measurement precision
If multiple machine learning models are used to analyze content, then selection accuracy is improved, but device complexity increases
Solution Approach 1:
The system segments the content analysis task across multiple specialized machine learning models, each responsible for detecting specific types of interesting moments or content characteristics. This segmentation allows for high selection accuracy through division of labor while managing complexity by organizing models into modular, independent units with clearly defined responsibilities
Data Source
AI summary
A data processing apparatus adapted to generate a recording for a content includes: input circuitry configured to obtain image data and corresponding audio data for the content, the image data including a plurality of image frames, a plurality of machine learning models including one or more first machine learning models and one or more second machine learning models, each first machine learning model configured to obtain at least a part of the image data and trained to output respective indicator data for the plurality of image frames, each second machine learning model configured to obtain at least a part of the audio data and trained to output respective indicator data for the plurality of image frames corresponding to the at least part of the audio data, where the respective indicator data associated with an image frame is indicative of an interest score for the image frame, authoring circuitry configured to select one or more groups of image frames from the plurality of image frames in dependence upon respective indicator data associated with the plurality of image frames, and recording circuitry configured to generate the recording for the content, the recording including one or more of the selected groups of image frames.