The AI Wire

EditYourself: Audio-Driven Generation and Manipulation of Talking Head Videos with Diffusion Transformers [TOP LAB](arxiv.org)

2026-02-01|paper|arXiv

Current generative video models excel at producing novel content from text and image prompts, but leave a critical gap in editing existing pre-recorded videos, where minor alterations to the spoken sc...

cs-CV cs-GR cs-LG

EditYourself: Audio-Driven Generation and Manipulation of Talking Head Videos with Diffusion Transformers [TOP LAB](arxiv.org)

2026-01-30|paper|arXiv

Current generative video models excel at producing novel content from text and image prompts, but leave a critical gap in editing existing pre-recorded videos, where minor alterations to the spoken sc...

cs-CV cs-GR cs-LG

SketchRodGS: Sketch-based Extraction of Slender Geometries for Animating Gaussian Splatting Scenes [TOP LAB](arxiv.org)

2026-01-06|paper|arXiv

Physics simulation of slender elastic objects often requires discretization as a polyline. However, constructing a polyline from Gaussian splatting is challenging as Gaussian splatting lacks connectiv...

cs-GR cs-CV

DiffusionBrowser: Interactive Diffusion Previews via Multi-Branch Decoders (arxiv.org)

2025-12-16|paper|arXiv

<think>

cs-CV cs-AI cs-GR

Moment-Based 3D Gaussian Splatting: Resolving Volumetric Occlusion with Order-Independent Transmittance (arxiv.org)

2025-12-15|paper|arXiv

<think>

cs-CV cs-GR

Native 3D Editing with Full Attention (arxiv.org)

2025-11-24|paper|arXiv

<think>

cs-CV cs-GR