Sports Video Codec With Unified Player Detection and Masking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional sports vision analysis systems are computationally redundant and time-consuming due to vertically structured modules that independently encode and decode visual information, and they lack the ability to support real-time object detection and foreground identification.
Innovation Solution
A unified machine learning architecture is introduced, combining object detection and foreground identification using a mask subnet attached to an object detection module, allowing for efficient real-time sports visual information extraction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vertically structured modules are used for sports vision analysis, then comprehensive visual information processing is achieved, but computational redundancy and processing time increase significantly
Solution Approach 1:
The patent combines multiple vertically structured modules (object detection, foreground identification, tracking) into a unified neural network architecture that processes visual information in a single forward pass, eliminating the need for sequential encoding and decoding operations while maintaining comprehensive analysis capabilities
Solution Approach 2:
The unified neural network architecture performs multiple functions simultaneously - object detection, foreground identification, and tracking - within a single model that processes the entire video stream in one operation, rather than requiring separate specialized modules for each task
2Adaptability or versatility
If multiple independent modules are used for object detection and foreground identification, then comprehensive analysis is achieved, but system complexity and computational overhead increase
Solution Approach 1:
The patent merges object detection and foreground identification into a single unified neural network model that processes visual information through one architecture, eliminating the need for multiple independent modules and their associated interfaces and data flows
Solution Approach 2:
The unified neural network is organized into functional components (input layer, processing layers, output layers) that work together in a coordinated manner, providing modular design benefits without requiring separate independent modules for each function
Data Source
AI summary
A computing system receives a broadcast video stream of a game. A codec module of the computing system extracts image level features from the broadcast video stream. The codec module includes an object detection portion configured to detect players in the broadcast video stream and a subnet portion attached to the object detection portion. The subnet portion is configured to identify foreground information of the detected players. The codec module provides the image level features to a plurality of task specific modules for analysis. The plurality of task specific modules generates a plurality of outputs based on the image level features.