I haven't had the chance to test the Corridor Crew team's latest CorridorKey yet, due to resource constraints, so I focused on what I could achieve with my 16 GB RTX 4070 Ti Super using ComfyUI. While the results for the Roto (U2net-humansegmentation) were not quite satisfying, I managed to achieve some interesting results by combining ControlNet and the Wan2.2 video model. Since I did this test there is an option to make it work with lower VRAM testing still in progress

1. GaussianSplat(Sharp) pointcloud used as TransitionMask, rendered in Houdini 2.Generated concept with ControlNet in ComfyUI FLUX 3.Additional Character with ControlNet generated in ComfyUI WAN22 #GaussianSplatting   hashtag#Houdini   hashtag#comfyui   hashtag#WAN22   hashtag#FLUX   hashtag#GenerativeAI   hashtag#CreativeTech   hashtag#SideFX

I’ve been experimenting with merging AI and VFX workflows for nearly 3 years now, and the progress is nothing short of incredible. My focus has been primarily on the ComfyUI – Houdini ecosystem. Thanks to the Houdini ComfyUI Bridge (developed by Rafael Drelich and Anatolii Iudanov), I’ve built a prototype to test the Trellis2 model—now fully integrated within Houdini. The goal was to Generate a Textured 3D Mesh, based on a single image. While automated segmentation is still a

ComfyUI Tests