Sublime

I’m releasing YOLO11 Object detection tutorial tomorrow. I’ll show you how to fine-tune YOLO11 on custom dataset, to detect figures, text and tables in arXiv papers. subscribe if you don’t want to miss the vid: https://t.co/h0IrNfkzxz https://t.co/ZiDYPeVd43

Image toolbox for cropping, filters, drawing, OCR, and image tweaks https://t.co/2obhkIeahF

🚀 Excited to introduce FlashSplat in our new ECCV paper on 3DGS Segmentation! We treat 3DGS segmentation as a linear program, solving it optimally and 50x faster! 💻HF page: https://t.co/jvKCUc8is5 🔗Code: https://t.co/by9mfrhaF6 #ECCV2024 #AI #GaussianSplatting https://t.co/bfw7KjG7NJ

ByteDance just dropped SA2VA: a new family of vision LMs combining Qwen2VL/InternVL and SAM2 with MIT license 💗 The models are capable of tasks involving vision-language understanding and visual referrals (referring segmentation) for images and videos ⏯️ take a look 🧶 https://t.co/TE8A2PEvg6

OCR toolkit https://t.co/ui7ScfoEFD

Github最近又开源了一个王炸的AI换脸技术ROOP, 他牛X的原因是它只需要一张照片就可以生成完成逼真的照片和视频，缺点就是这个东西对新手不太友好： 首先你必须得会用python，其次你必须安装git和ffmpeg Mac用户的话，就可以只用homebrew一键去安装 安装完成就简单了，一张照片就可以了 效果绝对惊艳

SAMURAI: Zero-shot visual tracking with motion-aware memory https://t.co/dou1YCtvac