top of page
F5W2cn3WIAEsDYN.webp

Fearlessly tapping the power of AI tools

​

Let's address the elephant in the room – AI is everywhere, and it's both exhilarating and a bit intimidating, all rolled into one. Like previous revolutions, we find ourselves at the crossroads of adaptation or being left behind.

​

In this case study, we delve into a variety of creative AI tools and examine how they've been seamlessly integrated into our creative pipeline. It's an exciting journey, and by embracing these innovations, we're staying ahead of the curve.

AI generated image.png

Step number one

Midjourney was used to generate the initial base image using some simple prompts.

booth back wall.png

Step number two

The image resolution was enlarged to a resolution suitable for an exhibition booth using Adobe Photoshop's AI-powered Neural Filters. 

Step number three

The image still needed a bit of work, luckily Adobe's AI-powered Firefly filled in some of the gaps

AI generated image background.png

Step number four

Moving away from Adobe, we next move to Runwayml.com. Using the original image and some prompts we can create an animation in seconds that would normally have taken hours to produce. With Runwayml, I was also able to smoothly time-stretch the footage at a much lower cost than using the Twixtor plugin. 

AI generated image background.png

Step number five

I wanted to shift the depth of field manually and change the strength of the blue effect. Luckily, Runway can create a depth map which can be used in After Effects with the plugin Bokeh Pro

Bearing in mind that this was just a proof of concept and it is still a bit rough around the edges, a unique animation was created from scratch in less than half a day.

Other AI tools currently being used

I have found ElevenLabs to be the most realistic text-to-voice tool for creating voice-overs and voice-cloning. Closely followed by Play.ht. I am also keeping a close eye on Revoicer, They show a lot of potential with regard to their emotion editing controls, but their voice base-lines still sound robotic.

​

In a similar fashion to Revoicer, I have been testing out SoundDraw, This is an AI music generator with great controls. A music for a video editor, animator, or soundscape artist.

​

Lastly, I used Move.ai to create some custom MOCAP with just a few iPhones. On a similar note, there is Wonder Dynamics, which can record the mocap, remove the subject, and comp in 3D CG characters. 

bottom of page