My inspiration was Lewis Carroll's novel Alice in Wonderland, a story that I wanted to adapt to modern times and use to reflect on digital technologies, video games, and the use of devices, and the way in which technology often manipulates people's emotions and desires. I thought that adapting Lewis Carroll's novel was a good opportunity to test tools and explore the possibilities of generative AI. One of the main challenges I set myself was to create long sequences without cuts, based on the possibilities offered by Kling's “first and last frame,” for example. In addition, I wanted to work on character consistency and create a realistic and animated aesthetic. I also wanted to test lip-syncing.
As for the workflow I followed, the procedure was as follows:
- Script, prompts, and narrative development with Claude
- Images with Midjourney
- Image editing with Google Nano Banana & Photoshop
- Videos with Kling & Google VEO 3
- Voice with Luma MiniMax audio
- Music and sound design with Epidemicsound
- I used many of these tools in Freepik
One of the main achievements, I think, was aesthetic, being able to create long sequences without noticeable cuts or assembled fragments, as well as the consistency of the character during the first and second parts of the short film, which have different aesthetics.
Built With
- claude
- epidemicsound
- freepik
- googlenanobanana
- kling
- midjourney
- photoshop
- premiere
Log in or sign up for Devpost to join the conversation.