view post Post 300 Reply Tencent released a new depth model that generates temporally consistent depth maps over videos ⏯️Model: tencent/DepthCrafterDemo: tencent/DepthCrafterPaper: DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videos (2409.02095)You don't need to input anything other than video itself, no need for optical flow or camera poses! 🤩
view post Post 2258 Reply I ported the hottest new shape-optimized SigLIP 🔥 merve/siglip-so400m-patch16-256-i18nif you don't want to wait for the next transformers release install transformers from my PR https://github.com/huggingface/transformers/pull/32938 and initialize SigLIP from there
New Depth Models Recent depth models Running on Zero 88 🦀 DepthCrafter a super consistent video depth model Running on Zero 159 🚀 Depth Pro Running on Zero 53 🚀 LOTUS Depth apple/DepthPro Depth Estimation • Updated 13 days ago • 2.24k • 286
BRAVE Models 🦁 Models mentioned in https://huggingface.co/papers/2404.07204 facebook/dinov2-large Image Feature Extraction • Updated Sep 6, 2023 • 223k • 42 google/flan-t5-xl Text2Text Generation • Updated Nov 28, 2023 • 134k • 465 google/siglip-large-patch16-384 Zero-Shot Image Classification • Updated 26 days ago • 8.86k • 5 google/vit-huge-patch14-224-in21k Image Feature Extraction • Updated Feb 14 • 55k • 18