Today in Generative Media

Humans prefer AI content; Runway and Getty team up; genAI and the future of content

• Humans Prefer AI-Generated Content, New Research Suggests (Forbes)

• Runway partners with Getty Images to build enterprise ready AI tools (RunwayML). Exclusive: Runway incorporates Getty Images into its AI generated video (Axios)

• Generative AI And The Future Of Content Creation (Forbes)

• Judge questions credibility of AI chief’s evidence in court battle with Getty Images (The Telegraph)

• AI is Not Creative Per the USCO and the Courts – And That’s a Good Thing (IP Watchdog)

• Readout Guidance: Learning Control from Diffusion Features (project page)

• Style Aligned Image Generation via Shared Attention (project page)

• VividTalk: One-Shot Audio-Driven Talking Head Generation Based on 3D Hybrid Prior (project page)

• FSGS: Real-Time Few-Shot View Synthesis using Gaussian Splatting (project page)

• GPS-Gaussian: Generalizable Pixel-wise 3D Gaussian Splatting for Real-time Human Novel View Synthesis (project page)

• MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model (project page) Demo on Replicate.com.

• VideoRF: Rendering Dynamic Radiance Fields as 2D Feature Video Streams (project page)

• Generative Powers of Ten (project page)

• SparseGS: Real-Time 360° Sparse View Synthesis using Gaussian Splatting (arXiv)

• HiFi Tuner: High-Fidelity Subject-Driven Fine-Tuning for Diffusion Models (arXiv)

• My friend got these very minimal sketches from an interior designer / architect. I put them in Interior AI v2 and got these renders back (X)