Jump to content

Your Ad Could Be Here


Open Club  ·  17 members  ·  Rules

Music Video

Yet more AI imagery


GregB

Recommended Posts

My latest (and likely my last) music video. It had to wait almost 3 years after the track's release in 2020 before the very recent advances in AI allowed me to generate the images I had had in my head when creating the song.

 

In the past I've used a lot of stock video, largely from https://artgrid.io/.  The main benefits have been:

  • high quality - framing, content, colour, resolution
  • multiple 'story' clips - allowing for consistency of character/location
  • videos clips, provided they are relevant, can fill a LOT of time in a music video!!
  • price (unlimited downloads and total freedom of usage)

 

HOWEVER, if given the choice, I much PREFER to use still images. I think photography is far more impactful and the eye is free to examine and absorb an unchanging image.  This remains so even with a slow steady zoom/pan of an image.  And recent AI can produce exquisite stills of tremendous surprise, beauty, detail and 'imagination'.

 

It'd be nice to delve deeper into the issue of stills vs film if anyone wishes to contribute their thoughts.

 

Greg

Edited by GregB
Link to comment
Share on other sites

The very new and free Dall-E 3 has been highly praised for it's ability to create words within images.  Prior to this, Dall-E - and also others like MidJourney - could only display gooble-de-gook scribbles.
 

My prompt was:

photo of a pretty blond woman in her twenties wearing a white t-shirt which says "The Flat White Album" written in large black capital letters.

 

The resulting lettering IS remarkably good and it follows contours of both the body AND the cloth folds, but not one of the three logos presented was spelled correctly!!


No doubt this will be fixed within weeks ... and then MidJourney will likely frog leap Dall-E with many new features.

 

Progress in AI image generation is absolutely astounding.

 

image.jpeg.0967a19a8392494a0d7a968a3adc3b71.jpeg

  • Like 1
Link to comment
Share on other sites

I tried it with DALL-E 2 and had a well generated female with worse spelling and wrong number of letter groups (I can’t call them words), so it isn’t a new thing. It has to be said I thought regurgitating words I gave it would have been easier than generating a new person. Who knew?

  • Like 1
Link to comment
Share on other sites

  • Recently Browsing   0 members

    • No registered users viewing this page.

Your Ad Could Be Here

Guests are always welcome...

but...

JOINING as a MEMBER (FREE) provides you with many benefits:

  • it is FREE
  • you will NOT be sent emails UNLESS you sign up for them
  • + you can interact with posts
  • you can create new Topics
  • you can directly message other members
  • you can seek critiques of your own work
  • you can offer critiques on the work of others
  • after a few posts you can post your own music and videos
  • have your songs/videos considered for Songstuff's official Playlists


  • Current Donation Goals

    • Raised $1,040
×
×
  • Create New...

Important Information

By continuing to use our site you indicate acceptance of our Terms Of Service: Terms of Use, our Privacy Policy: Privacy Policy, our Community Guidelines: Guidelines and our use of Cookies We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.