Blockout to Concept Art using AI (Research Blog)
Published
18-8-2026
Type of Work
Portfolio of Creative Works or Performance
Abstract
One of the challenges I have noticed among 3D artists and students is the difficulty of creating concept art, especially when drawing is not their strongest skill. This project explores how 3D blockouts and generative AI can be combined to help bridge that gap, allowing artists to have stronger control over their creative process without relying too much on drawing skills or complex prompt engineering.
The project involved reviewing existing literature and conducting a small practical experiment using 3ds Max, tyDiffusion and Stable Diffusion. A basic 3D environment was created using simple geometry, followed by several tests using ControlNet and IP-Adapter to explore how AI could be guided through 3D information. Different prompts, settings and scene elements were tested to understand how they affected the final outcome.
The experiment showed that using 3D blockouts can provide better control over composition, perspective and the overall design, while AI can help develop lighting, atmosphere and visual details. Although the results were not always perfect and required further refinements, the workflow provided a practical alternative for artists to explore and communicate their ideas.
The entire process was documented through a YouTube tutorial and a reflective blog. This project also helped me gain a better understanding of research methodology and how emerging AI technologies can be integrated into my creative practice and potentially support students in their concept art development.
This project explores how 3D blockouts and generative AI can be combined to help 3D artists develop concept art without relying heavily on traditional drawing skills or complex prompt engineering. The aim was to investigate a practical workflow that allows artists to have greater creative control while using AI to support the visual development process.
The research involved reviewing existing literature on generative AI, prompt engineering and controlled image generation, followed by a practical experiment using 3ds Max, tyDiffusion and Stable Diffusion. A simple 3D environment was created using basic geometry, with ControlNet and IP-Adapter used to guide the AI-generated results. Different compositions, prompts and settings were tested to observe how effectively the artist could control the final outcome.
The experiment demonstrated that combining 3D blockouts with AI can provide greater control over composition, perspective and visual direction while allowing artists to explore lighting, atmosphere and design variations more efficiently. Although the workflow still requires artistic knowledge and further refinement, it offers an alternative approach to concept-art development. The project was documented through a YouTube tutorial and a reflective blog, contributing to my ongoing creative practice and exploring how emerging technologies can support 3D artists and students.
References
Ko, H.-K., Park, G., Jeon, H., Jo, J., Kim, J., & Seo, J. (2023). Large-scale Text-to-Image Generation Models for Visual Artists’ Creative Works. Proceedings of the 28Th International Conference on Intelligent User Interfaces. https://doi.org/10.1145/3581641.3584078
Liu, V., & Chilton, L. B. (2022). Design guidelines for prompt engineering text-to-image generative models.
Oppenlaender, J., Linder, R., & Silvennoinen, J. (2025). Prompting AI art: An investigation into the creative skill of prompt engineering. International Journal of Human–Computer Interaction, 41(16), 10207–10229. https://doi.org/10.1080/10447318.2024.2431761
Zhang, L., Rao, A., & Agrawala, M. (n.d.). Adding Conditional Control to Text-to-Image Diffusion Models [Review of Adding Conditional Control to Text-to-Image Diffusion Models]. Retrieved July 13, 2026, from https://arxiv.org/pdf/2302.05543
Ibele, T. (2026). tyFlow (Version 1.114) [Computer software]. https://pro.tyflow.com/
Naughty Dog. (2013). The last of us (PlayStation 3 version) [Video game]. Sony Computer Entertainment.
Citation
Naval, A. (2026, October 30). Blockout to Concept Art using AI. Blogger. https://www.3damit.com/