{"id":335,"date":"2026-03-26T19:29:55","date_gmt":"2026-03-26T19:29:55","guid":{"rendered":"https:\/\/nicholasneff.com\/?page_id=335"},"modified":"2026-03-26T19:29:55","modified_gmt":"2026-03-26T19:29:55","slug":"blog-on-learning-more-about-ai","status":"publish","type":"page","link":"https:\/\/nicholasneff.com\/index.php\/blog-on-learning-more-about-ai\/","title":{"rendered":"Blog on Learning More About AI"},"content":{"rendered":"\n

This blog is about documenting my progress on learning what AI can do for image\/video generation, reimagining existing media, and integration pipeline workflows with softwares like Nuke, Maya, or Blender.<\/p>\n\n\n\n

<\/div>\n\n\n\n

Fighter x BAPE<\/h2>\n\n\n\n
\n
\r\n
<\/div>\r\n\r\n \r\n \r\n <\/div>\n\n\n\n
\n

I’ve wanted to tap into making work for brands, and through some brainstorming I found some pictures from Pintrest that I liked. Here’s what I found!<\/p>\n<\/div>\n<\/div>\n\n\n\n

\n
<\/div>\n\n\n\n
<\/div>\n<\/div>\n\n\n\n
\n
\n
\"\"<\/figure>\n<\/div>\n\n\n\n
\n

When I found the image that I liked the most, I used Higgsfield Soul 2.0 to regenerate a subject that matches the same attitude and look as the example. I learned to use key words, formats of spacing, positive and negative direction to have the AI model follow suit to what I wanted to create. This was the final result! With some more specific prompts and direction, I could direct the subject in any way I see for a better pose.<\/p>\n<\/div>\n<\/div>\n\n\n\n

<\/div>\n\n\n\n
\n
\n
\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n<\/figure>\n<\/div>\n\n\n\n
\n

This is the clothing set I had in mind. The brand is called BAPE, a premier Japanese streetwear brand founded by Nigo in 1993. It’s been a very recognizable brand of clothing, and I love the style, so I wanted my subject to wear this. The idea was to have a fighter who walks into a match wearing designer, like he’s promoting the brand just by showing up.<\/p>\n<\/div>\n<\/div>\n\n\n\n

<\/div>\n\n\n\n
\n
\n
\"\"<\/figure>\n<\/div>\n\n\n\n
\n

After doing my own research and development on which AI models are the best at what they do differently, I happen to find that Nano Banana, an AI model from Google (Gemini) actually does incredibly well at applying clothes to subjects. In Gemini, I submitted the two images, and got a result just like this; a clean, properly fitting, convincing image of my new subject wearing these new clothes. With this success, I moved onto turning this image into a small video for promotional purposes for the brand.<\/p>\n<\/div>\n<\/div>\n\n\n\n

<\/div>\n\n\n\n
\n
\n