Image Processing Apparatus Distance Map Selection for Relighting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional image processing techniques for correcting distance maps used in relighting processes are prone to errors due to noise and spots, and the suitability of distance information depends on the specific image processing task and type of virtual light source, making it challenging to achieve desired lighting effects.
Innovation Solution
An image processing apparatus that includes multiple acquisition units for obtaining distance distribution information based on parallax and feature quantities, with a determination unit to select the appropriate method for relighting, allowing for the generation of two different distance maps (sensor distance maps and feature point distance maps) to reduce errors and optimize lighting effects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single distance map acquisition method is used, then the device complexity is reduced, but the measurement precision and reliability of distance information deteriorate due to noise and spots
Solution Approach 1:
The distance map acquisition is segmented into multiple independent methods: a first acquisition unit that generates distance maps based on parallax information from multiple images, and a second acquisition unit that generates distance maps based on feature quantity analysis. Each unit operates independently with its own algorithm, allowing the system to divide the complex task of accurate distance measurement into manageable segments that can be evaluated and combined separately.
Solution Approach 2:
The determination unit serves as a universal selector that can choose between different distance map acquisition methods based on the specific relighting mode and image characteristics. This multi-functional unit enables the system to adaptively switch between parallax-based acquisition, feature-based acquisition, or combine both approaches, making the distance measurement system universally applicable to various relighting scenarios without requiring separate dedicated systems for each method.
2Measurement precision
If distance information is corrected for each pixel using clustering, then the measurement precision improves, but the productivity and calculation efficiency deteriorate due to high computational costs
Solution Approach 1:
Instead of applying computationally intensive clustering algorithms to every single pixel in the distance map, the system applies partial correction only to specific regions or uses simplified correction models for areas where high precision is less critical. This partial action approach maintains measurement precision in key areas while significantly reducing the overall calculation burden and improving processing efficiency.
Solution Approach 2:
The system dynamically changes parameters such as the correction threshold, processing resolution, and algorithm selection based on image characteristics and relighting requirements. For example, in regions with high noise but low visual importance, the correction parameter may be reduced or skipped entirely, while in critical regions like subject boundaries or high-contrast areas, full precision correction is applied. This parameter adaptation allows the system to maintain overall precision while improving productivity.
3Adaptability or versatility
If multiple distance maps are generated and selected based on relighting mode, then the adaptability improves, but the device complexity and processing time increase
Solution Approach 1:
The system performs preliminary evaluation of the input image characteristics and relighting mode requirements before generating distance maps. Based on this preliminary assessment, the determination unit pre-selects the most suitable acquisition method or combination of methods, avoiding the generation of unnecessary distance maps. This preliminary action significantly reduces processing time while maintaining adaptability to different relighting scenarios.
Solution Approach 2:
The system dynamically adjusts the distance map generation process based on real-time analysis of image characteristics and relighting mode. Rather than statically generating all possible distance maps, the system flexibly switches between acquisition methods and adjusts processing parameters according to the specific requirements of each relighting mode, optimizing the balance between adaptability and processing efficiency.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enables accurate and efficient generation of distance information with reduced calculation costs, allowing for precise control of lighting effects by selecting the appropriate distance map based on the relighting mode, thereby enhancing the quality of virtual lighting in images.
Implementation Method 1
a first acquisition unit configured to acquire the distance distribution information about the subject included in the image, based on parallax information about the image
Data Source
AI summary
An image processing apparatus includes a plurality of acquisition units configured to acquire distance distribution information about a subject included in the image, the plurality of acquisition units including a first acquisition unit configured to acquire the distance distribution information about the subject included in the image, based on parallax information about the image, and a second acquisition unit configured to acquire the distance distribution information about the subject included in the image, based on a feature quantity of the image different from the parallax information, and a determination unit configured to determine the acquisition unit which has acquired the distance distribution information with which a relighting process is to be performed, among the plurality of acquisition units.