Content-Aware Digital Asset Generation With ML-Based Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional digital asset generation systems are inefficient due to labor-intensive interactive procedures and require multiple graphical user interfaces and computational models, leading to significant turnaround time and resource consumption.
Innovation Solution
A machine learning model is utilized to detect and generate various digital assets from a digital image, including shape, color, pattern, and font assets, with an intelligent ranking system to recommend production-ready assets, reducing user interactions and computational overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If conventional digital asset generation systems use multiple separate graphical user interfaces and computational models, then different digital assets can be generated, but the operation becomes tedious and inefficient with significant turnaround time
Solution Approach 1:
The patent combines multiple separate computational models and graphical user interfaces into a single integrated system. The unified model simultaneously processes images to generate multiple asset types (stickers, emojis, clippings, GIFs) through one interface, eliminating the need for users to switch between separate tools and reducing operational complexity
Solution Approach 2:
The system implements a universal model that performs multiple functions - detecting objects, generating various asset types, and providing recommendations all within a single platform. This multi-functional approach allows one system to replace multiple specialized tools, improving ease of operation while maintaining versatility
2Ease of manufacture
If conventional systems provide tools for generating digital assets from the ground up or using templates, then custom assets can be created, but the process requires labor-intensive interactive procedures
Solution Approach 1:
The system performs preliminary object detection and asset generation automatically before user interaction is needed. By pre-processing images to identify objects and generate asset candidates in advance, the system reduces the time users need to spend on manual creation while maintaining custom asset quality
Solution Approach 2:
The system enables self-service asset generation by automatically detecting objects and creating multiple asset types without requiring users to manually configure parameters or navigate complex tools. The automated recommendation system selects appropriate assets based on detected content, allowing users to obtain custom assets with minimal interaction
3Adaptability or versatility
If multiple separate computational models are used to generate different digital assets, then various asset types can be produced, but resource consumption increases
Solution Approach 1:
The patent merges multiple separate computational models into a single unified model that handles object detection and multiple asset generation tasks simultaneously. This consolidation reduces redundant computations and resource consumption while maintaining the ability to generate diverse asset types including stickers, emojis, clippings, and GIFs
Data Source
AI summary
The present disclosure describes methods, systems, and non-transitory computer-readable media for implementing a machine learning framework to generate a recommend digital assets from a digital image. For example, in one or more embodiments, the disclosed systems utilize a machine learning model to detect a shape, color, pattern, or other digital asset type from a digital image and then extract (and further modify) the detected asset type to create various different digital assets as recommendations. In some cases, the disclosed system utilizes the machine learning model to determine one or more digital asset classes associated with the digital image, generate preprocessed digital assets from the digital image for those digital asset classes, and generate production-ready digital assets from the preprocessed digital assets. Further, in some instances, the disclosed systems provide one or more of the digital assets via recommendations based on asset scores determined via the generation process.