Today in Generative Media

What's next for gen video; object removal and insertion; style/content separation

• What’s next for generative video (MIT Technology Review)

• ObjectDrop: Bootstrapping Counterfactuals for Photorealistic Object Removal and Insertion (project page)

• Implicit Style-Content Separation using B-LoRA (project page)

• Design 3D apps and games faster than ever before with Bezi AI.

• Arc2Face: A Foundation Model of Human Faces (project page) Code on GitHub. Demo on HuggingFace.

• DesignEdit: Multi-Layered Latent Decomposition and Fusion for Unified & Accurate Image Editing (project page)

• DINO-Tracker: Taming DINO for Self-Supervised Point Tracking in a Single Video (project page)

• One2Avatar: Generative Implicit Head Avatar for Few-shot User Adaptation (project page)

• Garment3DGen: 3D Garment Stylization and Texture Generation (project page)

• AniPortrait: Audio-Driven Synthesis of Photorealistic Portrait Animation (arXiv) Code on GitHub. API on Replicate.

• ViTAR: Vision Transformer with Any Resolution (arXiv)

• Gamba: Marry Gaussian Splatting with Mamba for single view 3D reconstruction (arXiv)

• DN-Splatter: Depth and Normal Priors for Gaussian Splatting and Meshing (arXiv)

• Tutorial on Diffusion Models for Imaging and Vision (arXiv)

• We release the report of VideoCausalVAE, which supports both images and videos. The text-to-video model is on the way. (X)

• The very best in AI cinema is coming to Los Angeles on May 1. Request your ticket to the second annual AI Film Festival at http://aiff.com (X)