Original article excerpt
Server-side extracted preview paragraphs from the original source.
In this post, we explore how Outpost VFX achieved 8x faster training speeds using AWS infrastructure to transform their face replacement workflow, the technical architecture they implemented to overcome single-GPU limitations, and the measurable results achieved through AWS multi-GPU training.
AI model training for visual effects (VFX) can take weeks, creating bottlenecks in production timelines. For Outpost VFX, which operates studios across the UK, Canada, and India delivering high-end film and episodic content, every day of delay impacts client deliverables and project schedules.
In this post, we explore how Outpost VFX achieved 8x faster training speeds using AWS infrastructure to transform their face replacement workflow, the technical architecture they implemented to overcome single-GPU limitations, and the measurable results achieved through AWS multi-GPU training.
Traditional face replacement workflows in visual effects production require over 5 days of compositing or specialist beauty and de-aging support to create initial versions for director approval. While effective, these methods create bottlenecks early in the iterative approval process, the phase that is most critical to production timelines. For VFX professionals, slow AI training translates directly to missed deadlines, increased costs, and delayed client feedback cycles.
Outpost VFX had developed an AI model capable of training on on-set footage to accelerate face replacement processes. However, efficiency was constrained by single-GPU compute limitations. The existing face swap tool could only utilize one GPU at a time, limiting video random access memory (VRAM) access and processing capacity for model training operations. This prevented the team from realizing the full potential of their AI-assisted approach.
Outpost VFX identified three critical technical requirements for optimizing their AI workflow:
To address these requirements, Outpost VFX collaborated with AWS Generative AI Innovation Center developers who worked as an extension of their technology department to modernize their AI learning algorithms. The AWS Generative AI Innovation Center is a team of strategists, data scientists, engineers, and solutions architects that works step-by-step with customers to build bespoke solutions that harness the power of generative AI. Learn more about how to engage with the team on the Generative AI Innovation Center webpage.