What’s in a professional Photo?
As I embarked on my AI journey, I began by exploring the creative possibilities of generating digital artwork. While some might hesitate to categorize this as AI-generated ‘art’, I prefer to think of it as a collaboration between human creativity and machine intelligence. The process is indeed meticulous, requiring careful setup, experimentation with various parameters, and thoughtful curation of prompts to coax out visually appealing results. In many ways, the art of AI-generated imagery is not unlike traditional artistic practices – it demands attention to detail, an understanding of visual aesthetics, and a willingness to take calculated risks. Through this process, I've come to appreciate the unique strengths that AI brings to the creative table, and I'm excited to continue pushing the boundaries of what's possible in this space.
To gain hands-on experience with AI-generated art, I sought out localized software solutions that could help me understand the underlying technology. In my research, I stumbled upon stable diffusion, which surprisingly ran smoothly on modest hardware configurations, including as little as 4 GB of VRAM. The discovery of a localized program like this sparked my interest in learning more about the underlying technology, which led me to explore Windows CLI and GitHub. Through self-directed study, I gained a solid understanding of Python fundamentals. As I set up the necessary tools and frameworks, I began generating images and initially encountered mixed results. Undeterred, I dedicated countless hours to YouTube tutorials, experimentation, and iteration. After weeks of trial and error, I successfully refined my approach by leveraging better prompting techniques, utilizing tools like ControlNet and Segment Anything, and curating an extensive library of LoRas and checkpoints. This experience not only honed my technical skills but also deepened my appreciation for the art-science intersection.
As my skills evolved, I shifted my focus from creating aesthetically pleasing art to exploring ways to leverage AI-generated imagery for greater impact. I also sought to develop professional-grade work that would allow me to showcase my accomplishments on resumes, my portfolio, and social media platforms like Facebook and LinkedIn. Collaborating with Roop (now known as ReActor), I successfully generated images for this blog and various sections of my website. This experience further fueled my passion for AI, prompting me to explore new frontiers in the field, including NLP and LLMs. While opinions may vary on whether image generation is 'art' or not, for me, it marked the beginning of a thrilling journey through the world of AI.
