Combining AI-generated images with Blender, resulting in a very fast 3D test of style
A Glimpse into the Future
The realm of AI-generated images has seen remarkable advancements, with the quality of outputs escalating greatly each month. Pioneered by user-friendly services such as Midjourney, AI art is even making significant inroads into the broader public sphere. Motivated by a burgeoning interest in this evolving field, I delved into the domain by familiarizing myself with AI art tools, particularly Midjourney. My initial journey culminated in active participation and a prize-winning performance in a student AI art-themed hackathon hosted at NUS.
Learning the ropes
While Midjourney boasts impressive capabilities, its limitations in terms of minute customizations led me to explore alternative options. Exploring stable diffusion, I delved into technologies like ControlNet, AnimateDiff, and the powerful LoRA. My journey into the field of AI art took a more practical turn as I was fortunate enough to be hired part-time in a research unit dedicated to AI art at Osaka University. As a novice in this burgeoning field, I am captivated by the immense potential of these technologies and eagerly anticipate furthering my knowledge and skills.