The AI picture technology area is heating up in the meanwhile. Adobe has not too long ago launched the second iteration of Firefly, OpenAI launched Dall-E 3 with ChatGPT integration, and Midjourney has launched a 4X upscale of its photos.
Today, Meta AI is becoming a member of the picture technology occasion with the announcement of its very personal AI picture mannequin referred to as Emu.
I’ve seen that the preliminary picture outcomes of Emu launched by Meta are intently akin to these generated by Midjourney 5.2. So, on this article, I’ll examine the pictures from Emu and Midjourney utilizing related prompts.
Let’s get began.
Prompt: an afrofuturist girl carrying gold jewellery
Both AI fashions had been in a position to generate very intricate and extremely detailed photos. It now boils down to non-public desire. For me, I favor the picture on the fitting, generated by Emu.
Prompt: a playful kitten amusing itself with a yarn in a room bathed in daylight
Both cats look cute! But I believe the one on the fitting from Emu appears to be like extra real looking.
Prompt: a cool orange cat carrying sun shades taking part in a guitar with a gaggle of dancing bananas
Midjourney didn’t generate a coherent picture for this immediate. The characters within the background will not be bananas. So for coherency, Emu is superior right here.
Prompt: a horse studying a guide
Again, each photos are spectacular. For this one, I personally favor the whimsical fashion of the Midjourney picture.
