A new research paper introduces latent semantic planning as a technique to improve video diffusion models, targeting one of the core challenges in AI-generated video: producing ...
·
JoyAI has introduced Echo, a new model aimed at advancing long-form audio-visual content generation. The system targets two persistent challenges in generative media: maintainin...
WorldString proposes a new structured representation method for encoding actionable world states in AI systems. The approach aims to give AI models a more tractable way to descr...
Researchers from the University of Illinois have introduced Generative Blocks World, a system that lets users interactively edit generated images by representing scenes as assem...
·
NVIDIA Research's Spatial Intelligence Lab has unveiled ArtiFixer, a two-stage pipeline that addresses a persistent weakness in 3D Gaussian Splatting (3DGS): poor extrapolation ...
Researchers from Netflix and Caltech have published Vera, a layered diffusion model that jointly generates an edit layer, alpha matte, and composite video from a source clip and...
Researchers have introduced Go-with-the-Track, a single model that brings together motion control and video compositing—two capabilities that existing approaches treat as separa...
Researchers from Apple and the University of Hong Kong have introduced LGTM (Less Gaussians, Texture More), a new method for 3D Gaussian Splatting that scales to native 4K resol...
·
Researchers from Aalto University and ELLIS Institute Finland have introduced Smol-GS, a compression method for 3D Gaussian Splatting (3DGS) that dramatically reduces model size...
Researchers have introduced UniTemp, a single 1.3B-parameter autoregressive video model capable of generating video in any temporal direction—forward, backward, or interpolated ...
·
