Music Information Generating Device for Visual-Auditory Similarity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating music from pictures lack completeness, as they only utilize a part of music elements, resulting in incomplete music, and fail to maintain visual- auditory similarity, making it difficult to generate music appropriate for a picture-image and create a music score that can be objectively evaluated and revised.
Innovation Solution
A music information generating device that inputs a picture-image, divides it into lattice-shaped blocks, determines representative colors, and assigns sound sources based on color attributes, generating a music score that synchronizes with the visual elements, allowing for the creation of music that is more appropriate to the picture-image and can be objectively evaluated.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If music is generated only from RGB values at physical level, then conversion to music elements is simple, but visual-auditory similarity and music completeness deteriorate
Solution Approach 1:
The patent transforms color information from physical RGB values to perceptual color attributes (hue, saturation, brightness) that correlate with human visual perception. This parameter transformation enables the system to maintain visual-auditory similarity while still achieving automated music generation, resolving the contradiction between ease of manufacture and reliability.
2Ease of manufacture
If only one picture-image is processed, then processing is simple, but music slideshow capability and versatility deteriorate
Solution Approach 1:
The patent designs the music generation system to handle both single picture-images and sequences of picture-images using the same core processing logic. The system automatically detects whether input is a single image or a sequence, and generates either standalone music or synchronized music slideshows, achieving multi-functionality without complicating the basic processing framework.
3Device complexity
If music elements are partially utilized, then generation process is simple, but music completeness and quality deteriorate
Solution Approach 1:
The patent segments the music generation process into distinct functional modules: color attribute extraction, music element mapping, and score generation. Each module handles a specific aspect of the transformation, allowing the system to utilize multiple music elements (pitch, rhythm, timbre) comprehensively while keeping each segment's complexity manageable. This modular segmentation resolves the contradiction between process simplicity and music completeness.
Data Source
AI summary
A music information generating device, including: a block color name recognizing part which selects and determines, from among plural color names which are made to correspond to plural ranges of color attribute value having been set beforehand, a color name corresponding to the representative-color of the block as a block color name of the block, by recognizing to which range of the color attribute value the representative-color of the block belongs, where the plural color names have been made or are going to be made to correspond to the plural sound-source names having been stored; and a diagram music-score generating unit which selects, based on a criterion having been set beforehand, a designated block from among the blocks having been arranged in lattice shape, and thereby generates a diagram music-score.