Look, Compare and Draw: Differential Query Transformer for Automatic Oil Painting

TL;DR

This paper presents the Differential Query Transformer (DQ-Transformer), a neural model inspired by human painting processes, that improves automatic oil painting by focusing on incremental stroke impact, resulting in more realistic and expressive artworks.

Contribution

The introduction of the DQ-Transformer architecture that uses differential image analysis and adversarial training to enhance stroke prediction in neural oil painting models.

Findings

DQ-Transformer produces more refined and nuanced strokes.

The model achieves higher visual realism and artistic authenticity.

Fewer strokes are needed to produce high-quality paintings.

Abstract

This work introduces a new approach to automatic oil painting that emphasizes the creation of dynamic and expressive brushstrokes. A pivotal challenge lies in mitigating the duplicate and common-place strokes, which often lead to less aesthetic outcomes. Inspired by the human painting process, \ie, observing, comparing, and drawing, we incorporate differential image analysis into a neural oil painting model, allowing the model to effectively concentrate on the incremental impact of successive brushstrokes. To operationalize this concept, we propose the Differential Query Transformer (DQ-Transformer), a new architecture that leverages differentially derived image representations enriched with positional encoding to guide the stroke prediction process. This integration enables the model to maintain heightened sensitivity to local details, resulting in more refined and nuanced stroke generation. Furthermore, we incorporate adversarial training into our framework, enhancing the accuracy of stroke prediction and thereby improving the overall realism and fidelity of the synthesized paintings. Extensive qualitative evaluations, complemented by a controlled user study, validate that our DQ-Transformer surpasses existing methods in both visual realism and artistic authenticity, typically achieving these results with fewer strokes. The stroke-by-stroke painting animations are available on our project website.