Today in LS2, we used our previous experience with AI to create images. Our goal was to make pictures similar to a specific photo of a leopard lying on a tree branch, looking out. My partner and I each used different techniques and styles to make our images look unique. To do this, we had to give the AI a detailed description of what we wanted, including three main parts: the subject (what’s in the picture), the description (more details about the scene), and the style (how the picture should look artistically).
For example, I gave a prompt that explained the scene like this: “A leopard is resting peacefully on a tree branch in the jungle. The image should look like it was painted with watercolors, which gives it a soft, flowing, and artistic appearance, similar to watercolor paintings. The overall mood combines the calmness of the leopard with the gentle, artistic feel of watercolor art, making the picture look peaceful and beautiful.”
This helps the AI understand exactly what kind of image I want, including the scene itself and how it should look. We practice this process to get better at making creative and realistic images with AI.