Logo Detection Heat Map for Video Encoding Quality

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing video encoding systems face challenges in identifying and enhancing branding logos within video content without prior knowledge of their position, type, or structure, which affects the quality of logo representation in encoded video streams.

Innovation Solution

A logo detection system using a heat map with pixel accumulators to track and identify branding logos by analyzing multiple frames, determining their position and size without prior knowledge, and providing this information to the video encoding system to encode the logo sections with higher quality.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If prior knowledge of logo position, type, or structure is provided to the encoder, then logo encoding quality can be improved, but the system requires additional information input and preprocessing

Engineering Contradiction:
Improvelogo encoding qualityVSAvoidinformation input requirement
Core Design Contradiction:
Manufacturing precisionVSDevice complexity

Solution Approach 1:

The system performs self-detection of logo characteristics by automatically analyzing video frames to identify logo position, type, and structure without requiring external input of prior knowledge. The detector autonomously generates the information needed for high-quality logo encoding directly from the video content itself.

Inventive Principle:
Principle #25Self-service

Solution Approach 2:

The logo detection and characterization process is performed in advance before the encoding stage. By detecting and analyzing logo properties in prior frames, the system prepares the necessary information (position, type, structure) beforehand, enabling the encoder to apply appropriate quality settings without real-time complexity.

Inventive Principle:
Principle #10Preliminary action

2Adaptability or versatility

If logo detection is performed without prior knowledge of logo characteristics, then system adaptability is improved, but detection precision and reliability deteriorate

Engineering Contradiction:
Improvedetection system flexibilityVSAvoidlogo detection accuracy
Core Design Contradiction:
Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system transitions from analyzing single-frame static images to examining multi-frame temporal sequences. By accumulating and comparing logo characteristics across multiple frames, the detector enhances detection precision through temporal dimensionality, allowing accurate identification even without prior logo knowledge.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Solution Approach 2:

The system employs iterative detection and verification processes where initial detection results are refined through feedback from subsequent frame analyses. The detector continuously adjusts its identification based on accumulated evidence from multiple frames, improving accuracy while maintaining adaptability to unknown logo types.

Inventive Principle:
Principle #23Feedback

Data Source

PatentEP3227831B1Automatic detection of logos in video sequences
Publication Date: 2024.05.15 COMMSCOPE UK LTD
  • EP3227831B1 patent drawingFigure 1
  • EP3227831B1 patent drawingFigure 2A
  • EP3227831B1 patent drawingFigure 2B

AI summary

Particular embodiments automatically identify and track a logo that appears in video content. For example, particular embodiments can track a branding logos position and size without any prior knowledge about the logo, such as the position, type, structure, and content of the logo. In one embodiment, a heat map is used that accumulates a frequency of short-term logos that are detected in the video content. The heat map is then used to identify a branding logo in the video content.