Researchers from TU Munich have introduced GaussianGPT, an autoregressive transformer model that generates 3D Gaussian scenes through next-token prediction — offering an alterna...
·
Researchers Cao Duy and Phong Nguyen-Ha have introduced PixGS, a single-stage diffusion pipeline for generating 3D Gaussian Splats (3DGS) directly from text or images. Unlike pr...
·
Researchers from the Technical University of Munich, AUDI AG, and the University of Virginia have introduced TriFlow, a generative method for producing compact, artist-like 3D m...
·
Fal.ai has published a technical breakdown of how it reduced Ideogram v4 image generation at 1K resolution from 2.75 seconds down to 0.44 seconds — a 6× speedup — with no visibl...
·
MyWorld, a £70m six-year R&D programme funded by UKRI's Strength in Places Programme and focused on the West of England's creative economy, is advocating for structured "innovat...
SwitchHDR is a new research model designed to recover genuine 16-bit HDR from ordinary 8-bit SDR video. Rather than stretching the existing signal — which exposes banding and am...
·
A new research method called "One Video, One World" enables the reconstruction of physical 4D scenes from ordinary monocular (single-camera) video footage. The approach allows d...
A new market research report from Research and Markets forecasts the generative AI in movies sector will grow from $0.4 billion in 2025 to $1.03 billion by 2030, compounding at ...
·
A wave of studio AI partnerships and acquisitions is putting traditional animation and VFX roles under acute pressure. Lionsgate partnered with Runway to generate footage from m...
A wave of studio AI partnerships and acquisitions is putting traditional animation and VFX roles under acute pressure. Lionsgate has partnered with Runway to generate footage fr...
