Video Processor Architecture for Frame Rate Conversion

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current frame rate conversion techniques in video processing face challenges in real-time analysis and latency due to independent operation of frame rate conversion and other video processing functions, particularly when converting between different frame rates for enhanced video quality.

Innovation Solution

A method where video attribute data is processed upstream and passed through an auxiliary channel to a frame rate converter, allowing it to avoid re-analyzing video frames and reduce latency, involving a video processor with de-interlacing and attribute formatting, and a frame rate converter with interpolation based on received attribute data.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If frame rate conversion is performed independently from other video processing functions, then the frame rate converter can operate autonomously, but it must re-analyze video frames which increases latency and processing time

Engineering Contradiction:
Improveautonomous operationVSAvoidlatency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The video processor performs preliminary analysis of video frames and generates attribute data (motion vectors, scene change detection, interlaced/progressive flags) before the frame rate conversion occurs. This pre-computed attribute data is passed to the frame rate converter through an auxiliary channel, eliminating the need for the converter to re-analyze frames and significantly reducing processing latency.

Inventive Principle:
Principle #10Preliminary action

2Measurement precision

If the frame rate converter re-analyzes video frames to obtain attribute data, then it can ensure data accuracy, but this increases processing complexity and time

Engineering Contradiction:
Improveattribute data accuracyVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges the video processing and frame rate conversion functions into an integrated system where the video processor and frame rate converter operate cooperatively. The video processor generates attribute data that is passed through an auxiliary channel to the frame rate converter, allowing both functions to share processing resources and information while maintaining accurate attribute data without requiring duplicate analysis.

Inventive Principle:
Principle #5Merging (Combining)

3Productivity

If frame rate conversion is performed in real-time, then video quality enhancement is achieved, but the requirement to analyze incoming video increases processing demands

Engineering Contradiction:
Improvereal-time conversion capabilityVSAvoidprocessing power
Core Design Contradiction:
ProductivityVSPower

Solution Approach 1:

By performing video analysis and generating attribute data in advance during the video processing stage, the system prepares all necessary information for frame rate conversion before real-time conversion is needed. This preliminary action reduces the processing power required during real-time operation, as the frame rate converter only needs to apply pre-computed attributes rather than performing full frame analysis.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP2127371B1Video processor architecture and method for frame rate conversion
Publication Date: 2012.08.15 BROADCOM INC
  • EP2127371B1 patent drawingFigure 1~2
  • EP2127371B1 patent drawingFigure 3~4
  • EP2127371B1 patent drawingFigure 5

AI summary

A video processor, upstream of a frame rate converter determines video attribute data. This attribute data is formatted and passed along a channel to the frame rate converter. The frame rate converter extracts the attribute data from the channel for use in frame rate conversion. The frame rate converter may thus rely on attribute data obtained by the video processor, and need not re-analyze video frames.