2026 showreel
Directred & created the new brand commercial for the recently re-launched ‘Commodore’ brand.
Gen AI tests – It’s amazing how fast things change
This is an example of how fast things move. As I was doing this test, a new tool appeared that would dramatically make this process easier. By the time I downloaded it, there was an even newer tool that simplifies the process even more radically.
This test is a combination of creating a more robust LORA of myself and then swap out the background whilst keeping the main look of ‘me’ the same. Note the lighting changes from the red lights etc that appear on the face in a natural way.
It’s always good to use these these tools as quick sketches, knowing that the entire workflow will be different the next time !
This test was to explore the idea of retaining the visual consistency of on-screen talent that is potentially changed during ‘Style Transfer’ or other V2V methods, by creating a LORA of the same actor. This would be added during the generative time. This method would hopefully gain any lighting cues that only exist from the generative model.
So, First I made a bit more robust ‘LORA’ with approximate 80 images.
Rather than film a simple static scene, I opted to create a scene that might be more typical in production – something with movement and showed various scales of the person.
Although low res, This test retained the exact camera move and it does keep consistency on the face, note the body clothing has been changed completely, leaving the face with a more consistent look to the original.
Instead of relying on the Ai model to create the correct camera movements. I shot them on my phone, and used a combination of tools to swap the background and then me out of the shot.
I then did another tests that retained the same camera movement, but kept a more realistic submarine environment.
The last tests gained the benefit of using the newer models that have recently been released. I reused the LORA here, and used Veo2. While Veo3 is amazing, Veo2 has a better toolset for now. I made 2 versions, one only retaining the head, and the other a complete body swap.
Please ignore my sudden weight gain in this shot.
Conclusion
Although the test was hampered by lack of V-Ram, and these are simple tests, it’s clear that LORAs are quite transportable, and they do help to re-inforce the consistency of a live action character , and also gain lighting cues during the generative process..
Gen AI short film – Time after Time
This little idea was done in a few days and started life as only music. My wife and I speculated on the name of the track and came up with ‘Netflix fantasy drama, quirky girl finds mysterious clock in attic’, which let to this story of a woman who cheats time by visiting this age reversing clock. We see her revisitng this old friend of hers, again.
I used a combination of Ai tools to create this. ChatGPT vision model, Magnifica, Kling and Runway.
When doing these kinds of tests, the goal for me is to explore the tools, it’s never to have a perfect piece. I feel there’s a value in using and trying everything and see where it takes me. Using the ‘front door’ approach without using ComfyUI is becoming an interesting balance. Every week there is an update to the consumer tools that could often take a lot longer to set up in ComfyUI. We live in interesting times!
So, first off I created various looks of this girl at various ages in life, and also created contact sheets to aloow quick trianing.
I also created various ages of ‘Red Station Wagons’ that she would drive.
Instead of relying on the Ai model to create the correct camera movements. I shot them on my phone, and used a combination of tools to swap the background and then me out of the shot.
Whilst using ‘Comfy Ui’ would have resulted in a more consistant output, I wanted to do this test over a few days, so I went with the more typical ‘front door’ approach with Kling.
Whilst the images are all Ai, I did create the music in a non-Ai way, just sitting in my studio.
For some of the more cinematic shots I wanted a specific camera movement, as I could have spent time re-generating shots in AI describing with text, but instead I made a very basic house and forest, and animated the camera exactly how I wanted it. I didn’t need to spend any time on details as I’d give the AI model some other reference as stills. However, it does take the circle window from the attic.
This balance of using CG to get the camera angles, shot proportions,and the right pace is pretty efficient.
My original and longer edit had a shot where we see a little montage of her visiting the house when she was younger. In this version you can see a much younger version of her climbing the stairs. I added a few ‘younger’ versions of her exploring the attic. However, ultimately it felt a bit clunky and superfluous, but here it is.
Great talk with young creators at Cal State Fullerton
Had a wonderful time talking to students at California State University, Fullerton, with Shelley Armstrong Murray about the games industry in the 90’s. Yes, more Worms !
We had a diverse set of your creators, some game coders, some focusing on fine art, and some musicians too. It was refreshing to see such energized young artists !
Had a fantastic time at the Academy of the Arts & Sciences, helping my good friend Paul Debevec demonstrate the ‘Sodium Vapor Matting Process’ used to incredible effect in the movie ‘Mary Poppins’.
Demoing for the first time, a real-time sodium vapor matting technique. I helped putting the solution together in OBS along with making a custom shader to do the correct math, and it worked REALLY well !
Video to AI tests
A Client wanted to see the concept of using normal footage and blending into and AI world. I quickly made some ideas.