This weekend in Generative Media

Hollywood secretly uses AI; "AI On the Lot" conference; Fake ISIS news anchors

• Hollywood at a Crossroads: “Everyone Is Using AI, But They Are Scared to Admit It” (Hollywood Reporter)

• AI In Hollywood Draws Heavy Crowd For New Kind Of Filmmaking (Forbes)

• These ISIS news anchors are AI fakes. Their propaganda is real. (Washington Post)

• Woman who ‘saved’ Brooklyn Bridge gives AI-generated grad speech (New York Post)

• AI as a tool for storytellers: A conversation with Ed Catmull (YouTube)

• SimpleRecon: 3D Reconstruction Without 3D Convolutions (GitHub)

• Controllable and Interactive 3D Assets Generation with Proxy-Guided Conditioning (SIGGRAPH 2024 project page)

• Factorized Motion Fields for Fast Sparse Input Dynamic View Synthesis (SIGGRAPH 2024 project page)

• Analogist: Out-of-the-box Visual In-Context Learning with Image Diffusion Model (project page)

• LogoMotion: Visually-Grounded Code Generation for Content Aware Animation (project page)

• Text-to-Vector Generation with Neural Path Representation (project page)

• SOEDiff: Efficient Distillation for Small Object Editing (project page)

• Grounded Compositional and Diverse Text-to-3D with Pretrained Multi-View Diffusion Model (arXiv)

• 🤩3D Generation Arena🤩 Lots of 3D generation models come out recently. But which one is preferable from human perception? (X)

• Super practical AI workflow. Make a video using text/image to video. Then use depth estimation to understand the 3D scene structure, and add a subtle camera move. (X)

• audio-reactive ai animation (sound on) (X)