GenAI and Image Segmentation to spice up Presentations
I have been a satisfied user of generative image models such as stable diffusion, Midjourney and DALL·E 2 for over one year now. Admittedly, around 2016 or 2017, when the outputs of Generative Adversarial Networks (GANs) looked super pixelated and weird at times1, I did not get the appeal. Especially when compared with discriminative models that were more efficient and performed way better for sentiment classification and so on. However, like many people, I realized the great potential of generative approaches at the latest after Midjourney V3 and V4 were launched at the end of 2022....