VividSky is a desktop application for scientific image characterization, providing an accessible interface for exploring and exporting the structured visual data produced by the VividSky Python analysis engine. Rather than relying on subjective interpretation, the system applies algorithmic image analysis and computational techniques to measure objective visual properties of artwork, transforming images into structured datasets suitable for research, archival, and production workflows.
Developed as part of my digital catalogue raisonné initiative, VividSky was created to support the analysis of large collections of artwork by extracting consistent visual metrics that can be searched, compared, and studied over time. The system allows visual characteristics to be indexed alongside traditional catalogue information, making it possible to discover relationships between works, identify recurring color usage, group visually similar paintings, and explore broader patterns across an artist's body of work.
Scientific Image Characterization
VividSky extracts a wide range of measurable visual characteristics from artwork using computational image analysis and statistical methods. These include dominant color palettes, average color, color distributions, color temperature, brightness, saturation, contrast, and additional visual descriptors derived from algorithmic analysis. Where appropriate, lightweight machine learning techniques are incorporated to assist in feature extraction and classification, while the overall system remains grounded in deterministic, reproducible measurement rather than generative or interpretive AI.
Data Representation and Export
The underlying analysis engine produces structured data representations describing the measurable visual properties of each image. These representations can be exported in machine-readable formats for use within catalogue systems, research projects, archival databases, and custom production pipelines. By converting visual characteristics into structured metadata, large image collections become searchable, filterable, and suitable for computational analysis beyond what is practical through manual observation alone.
Desktop Application
VividSky provides a desktop interface that makes the analysis engine accessible without requiring users to work directly with Python scripts or structured data formats. Individual images or entire directories can be analyzed through the application, with extracted visual metrics presented through an organized interface for inspection, comparison, and export. The application serves as a practical wrapper around the underlying analysis engine, allowing artists, researchers, archivists, and collections to integrate computational image characterization into everyday workflows.
VividSky continues to evolve alongside my ongoing catalogue raisonné research, expanding both the range of measurable visual descriptors and the computational techniques used to characterize artwork while maintaining a focus on objective, reproducible visual measurement.
Availability
VividSky is currently available as part of my Catalogue Raisonné Digital Artifact service or through licensing. A public release is currently under development and will be announced at a later date.