Clean Energy Emitter

Assist. Prof. Dr. Gomesh Karnchanapayap

Introduction

Area of machine assisted art making is considered a cutting-edge way of creative exploration. This project explores the concept of how clean energy can be produced from the balance of using natural resources through animated artwork by using machine assisted art making process or Artificial Intelligence art (A.I. art). There are a variety of ways to make A.I. art, including procedural ‘rule-based’ picture production utilizing mathematical patterns, algorithms that replicate brush strokes and other painted effects, and artificial intelligence or deep learning methods like generative adversarial networks and transformers (Szegedy et al, 2014). In this study, the artist chose Contrastive Language-Image Pretraining (CLIP) which is considered an innovative method for creative exploration to create the creative result. CLIP is based on extensive research on zero-shot transfer, natural language supervision, and multimodal learning allowing machine to recognize a wide variety of visual concepts in images and associate them with their text (OpenAI.com, 2021). One of emerging CLIP platforms is Disco Diffusion
which runs on Google Collab (Disco Diffusion, 2022). By utilizing the text prompt “A beautiful painting of Thailand Grand Temple on a forested hill emitting green energy particles, the sky is full of smoke and cloud in the style of john Trending on artstation.” along with code settings feed to Disco Diffusion, a series of work in progress still paintings were produced showing steps machine interpreted and computed. The result is a mesmerizing painting steps in the mood and tone of the given text directive showing green particles emitting off trees along with Thai style temple. This technique should be further explored and utilized for in the future machine might take a larger role in producing creative works.

Conclusion

The result is a mesmerizing painting steps in the mood and tone of the given text directive showing green particles emitting off trees along with Thai style temple. This technique proves machines can be used to assist how humans can creatively explore ideas. In the future, the author believes computers will have a greater impact to art making process.

Objectives Aims or Purposes

– To study the concept of art exploration with Contrastive Language-Image Pre-training.
– To define proper text prompt for art creation with Contrastive Language-Image Pre-training.
– To create machine assisted artwork in the theme of “Clean Energy Emitter”

Process or Methods

– Studied and analyzed the concept of creating artworks by using Contrastive Language-Image Pre-training.
– Defined the text prompt to feed to the machine. “A beautiful painting of Thailand Grand Temple on a forest hill emitting green energy particles, the sky is full of smoke and cloud in the style of john Trending on artstation.”
– Utilized Disco Diffusion to begin the creation process with the following settings

Techniques and materials

Digital Animation creation through the use of Contrastive Language – Image Pre-training. to construct sequence of work in the progress paintings. The step paintings were then combined to create animation.