Today in Generative Media
Will Smith eats spaghetti fr; a Craigslist for GPUs; video captions and understanding
Will Smith parodies viral AI-generated video by actually eating spaghetti (Ars Technica)
We made a craigslist for gpu clusters: http://gpulist.ai (X)
Video ReCap: Recursive Captioning of Hour-Long Videos (project page)
VideoPrism: A Foundational Visual Encoder for Video Understanding (arXiv)
MVDiffusion++: A Dense High-resolution Multi-view Diffusion Model for Single to Sparse-view 3D Object Reconstruction (project page)
DiLightNet: Fine-grained Lighting Control for Diffusion-based Image Generation (arXiv)
Text-to-image after experiencing something like this? No way! (X)