I tested Dall.e 2, an AI that creates images from text

  • Thread starter Thread starter fluidistic
  • Start date Start date
  • Tags Tags
    Ai Images Text
Click For Summary

Discussion Overview

The discussion revolves around the effectiveness of AI image generation tools, specifically Dall.e 2 and Defusion Bee, in creating images from text prompts. Participants share their experiences and frustrations with the quality of the generated images, exploring the limitations and challenges of using these technologies.

Discussion Character

  • Exploratory
  • Debate/contested
  • Technical explanation

Main Points Raised

  • One participant expresses disappointment with Dall.e 2, describing it as "absolute garbage" and suggesting it is significantly inferior to ChatGPT.
  • Another participant shares a similar experience with Defusion Bee, noting that it produced results comparable to Dall.e 2 and mentioning that other users on DeviantArt seem to achieve better outcomes with different tools.
  • A third participant highlights the extensive effort required to achieve satisfactory results, citing a winner of an art competition who spent 200 hours refining prompts.
  • One participant argues that AI cannot create art, positing that the creative process is inherently human, regardless of the tools used.
  • Another participant recounts their experience configuring stable diffusion software and generating a large number of images, concluding that the results are mostly unsatisfactory at this stage.

Areas of Agreement / Disagreement

Participants generally express dissatisfaction with the current capabilities of AI image generation tools, indicating a consensus on their limitations. However, there is disagreement regarding the nature of art creation and the role of AI in that process.

Contextual Notes

Participants mention various tools and their experiences without resolving the effectiveness of each. The discussion reflects a range of personal experiences and subjective evaluations of the technology.

Who May Find This Useful

Individuals interested in AI-generated art, technology enthusiasts, and those exploring the capabilities and limitations of AI in creative fields may find this discussion relevant.

fluidistic
Gold Member
Messages
3,932
Reaction score
283
So I tested ChatGPT's cousin, Dalle.e 2. It is said to create images from text, and the website suggest to be very explicit in the description of the image, add many details, to get a good picture.
My mind started to wander... Would I finally get a pic of Einstein battling dinosaurs, or having a dinner with Newton while Trump and Putin are playing chess under the table? Boy did I was disappointed. Terribly disappointed.

Don't waste your time... this software is absolute garbage (for now at least), it's light years behind chatGPT.

Albert Einstein having a dinner with Isaac Newton while dinosaurs are playing chess under the table.
DALL·E 2023-02-13 22.15.38 - Albert Einstein having a dinner with Isaac Newton while dinosaurs...png
Music flowing in a river with pink reflections with physics equations on the water
DALL·E 2023-02-13 22.18.12 - Music flowing in a river with pink reflections with physics equat...png
Humanized apes having dinner on a table with Einstein while playing Bitcoin
DALL·E 2023-02-13 22.19.22 - Humanized apes having dinner on a table with Einstein while playi...png
A neandertal fighting an angry tyranosaurus rex with a screwdriver on a busy road
DALL·E 2023-02-13 22.20.40 - A neandertal fighting an angry tyranosaurus rex with a screwdrive...png
 
Reply
  • Haha
Likes   Reactions: mcastillo356
Computer science news on Phys.org
I had the same experience with Defusion Bee. It was about the same as what you are showing. People on DeviantArt are producing wildly better things with Ai, including complex stories with consistent appearances of the characters. I don't know what tools they are using, but it isn't DeviantArt's own Ai. The thing threatened to ban me for my first request, and didn't even say why.
 
Reply
  • Like
Likes   Reactions: Office_Shredder
The person who won that art competition said they spent 200 hours trying prompts before getting what they wanted. This isn't a fire and forget tool.
 
Reply
  • Like
Likes   Reactions: Algr
Yes. Ai can't create art. But then neither can pencils, cameras, or photoshop. Humans create art.

cosmic_dreamer__a_girl_s_journey_through_the_stars_by_oanarinaldi_dfoh951-pre.jpg

p2u_city_background_56_by_luffassets_dfnpb5y-pre.jpg
pretty_library__made_with_ai__by_kroniksan_df9fsg9-fullview.jpg
 
I have spent several evenings trying to configure stable diffusion to run on my computer using CUDA. I have spent even more hours generating over 24k images from different prompts (OK, my computer did, choosing and entering a prompt is like half a minute). I have but a few images that are reasonably close to being worth of showing.

Yes, at this stage it is mostly useless garbage.

1676362501600.png


1676362513156.png
 
Reply
  • Like
Likes   Reactions: BillTre, Algr and DennisN

Similar threads

Current and Future AI Threats to Humanity and the Human Response
  • · Replies 10 ·
Replies
10
Views
5K