AI, Kya Bolti Tu?

My experiments with AI stuffses

AI is blowing up right now, and it's only getting better. I'm trying to document my experiments with AI in this new series of posts. Come along for the ride as I hope to explore the possibilities of this amazing technology.

Part 1: A snapshot of Dall-E 2's progress over past few months using the same prompts


✒️prompt: a stained glass window depicting a raccoon eating a hamburger

Dall.E 2, ~Sep/Oct 2022
Dall.E 2, July 2023

✒️prompt: a surrealist painting by Salvador Dalí of a mouse sitting in a rollercoaster, digital art

Dall.E 2, ~Sep/Oct 2022
Dall.E 2, July 2023

✒️prompt: post apocalyptic futuristic city, cozy cafe, street, overhead train track, digital art, trending on artstation, hyper realistic, epic, india, night, aurora borealis, wide angle

Dall.E 2, ~Nov 2022
Dall.E 2, July 2023

✒️prompt: a sparrow wearing a helmet flying on a cloudy day, digital art

Dall.E 2, ~Oct/Nov 2022
Dall.E 2, July 2023

✒️prompt: a beautiful dreamcatcher in an aesthetic room, 24mm, bokeh, professional photography

Dall.E 2, ~Oct/Nov 2022
Dall.E 2, July 2023

Observations

  1. The way language models interpret prompts still blows my mind.

  2. The visual quality of images generated by language models hasn't really changed much in recent years.

  3. This could be because the prompts I used were pretty basic.

  4. To get better quality images, you usually need to craft your prompts more carefully.

Write a comment ...

Write a comment ...

Ishan Manjrekar

Designs Games | Makes Podcasts | Clicks Pictures | Travels | Curious