Moonworks Lunara Aesthetic II: An Image Variation Dataset

TL;DR

Lunara Aesthetic II is a new dataset of 2,854 image variation pairs designed to evaluate and improve contextual consistency, identity preservation, and aesthetic quality in image generation and editing systems.

Contribution

The paper introduces Lunara Aesthetic II, a publicly available dataset with controlled, identity-preserving image variations for benchmarking and training image generation models.

Findings

High identity stability in variations

Strong realization of target attributes

Superior aesthetic scores compared to web datasets

Abstract

We introduce Lunara Aesthetic II, a publicly released, ethically sourced image dataset designed to support controlled evaluation and learning of contextual consistency in modern image generation and editing systems. The dataset comprises 2,854 anchor-linked variation pairs derived from original art and photographs created by Moonworks. Each variation pair applies contextual transformations, such as illumination, weather, viewpoint, scene composition, color tone, or mood; while preserving a stable underlying identity. Lunara Aesthetic II operationalizes identity-preserving contextual variation as a supervision signal while also retaining Lunara's signature high aesthetic scores. Results show high identity stability, strong target attribute realization, and a robust aesthetic profile that exceeds large-scale web datasets. Released under the Apache 2.0 license, Lunara Aesthetic II is intended for benchmarking, fine-tuning, and analysis of contextual generalization, identity preservation, and edit robustness in image generation and image-to-image systems with interpretable, relational supervision. The dataset is publicly available at: https://huggingface.co/datasets/moonworks/lunara-aesthetic-image-variations.