Pinned
🎉 [𝗡𝗲𝘂𝗿𝗜𝗣𝗦 𝟮𝟬𝟮𝟱] 𝗣𝗮𝗽𝗲𝗿 𝗔𝗰𝗰𝗲𝗽𝘁𝗲𝗱 𝗮𝗻𝗱 𝗖𝗼𝗱𝗲 𝗥𝗲𝗹𝗲𝗮𝘀𝗲!
Not everyone is a photo pro — even with advanced AI tools. Restoring an image to professional grade often requires either 𝘥𝘦𝘦𝘱 𝘱𝘩𝘰𝘵𝘰𝘨𝘳𝘢𝘱𝘩𝘺 𝘦𝘹𝘱𝘦𝘳𝘵𝘪𝘴𝘦 or 𝘣𝘳𝘰𝘢𝘥
00:00
Zhengzhong Tu
1,289 posts
Zhengzhong Tu
@_vztu
- Reinforcement learning explained x.com/i/status/17310…
- State-of-the-art human pose estimation model x.com/isibhakss/stat…
- Dear NeurIPS reviewers, please be reminded to delete the GPT prompts next time :)
GIF- 📍 𝗖𝗮𝗻 𝗔𝗜 𝗡𝗮𝘃𝗶𝗴𝗮𝘁𝗲 𝗠𝗮𝗽𝘀 𝗟𝗶𝗸𝗲 𝗛𝘂𝗺𝗮𝗻𝘀 𝗗𝗼? 𝗜𝗻𝘁𝗿𝗼𝗱𝘂𝗰𝗶𝗻𝗴 𝗠𝗮𝗽𝗕𝗲𝗻𝗰𝗵! 🗺️🤖 𝘙𝘦𝘢𝘥𝘪𝘯𝘨 𝘮𝘢𝘱𝘴, like Google Maps and Theme Park Maps, is second nature for humans. It is a highly challenging task that requires visual understanding,
- Senior Engineer walking you through the production code x.com/uoldguy/status…
- "What a week, huh?" "Captain, it's Monday morning!" --- Happy to share that we've open-sourced the model weights and Colab demo of our #ECCV2022 paper "MaxViT: Multi-Axis Vision Transformer", finally! 🤗Pls check it out and play around with it: github.com/google-researc…
- Today, I "graduated" from Google. 今天，我从谷歌“毕业”了 👨🎓 📢 [TL;DR] I have closed an incredible chapter at #GoogleResearch and will be appointed at #CS Department Texas A&M University (@TAMU) as an AP—though I prefer "AI Professor" to Assistant Professor. 🌟 We're now
- 🚀🔥 New Paper: DINO-R1 — Bringing Reinforcement-Learned Reasoning to Pure Vision Models 🧠👁️ We’re excited to share DINO-R1, our new approach that brings reasoning into pure vision models using reinforcement-style training—an area that has largely been missing compared to the
- Overleaf's server after processing 30K NeurIPS papers + 11K ICCV rebuttals at the same time:
- 🚨Vision-Language Models (VLMs) are truly amazing. Ever wonder if their visual and textual "brains" always agree? I am excited to share our latest paper, where we tackle a critical challenge in VLMs, dubbed the 𝐜𝐫𝐨𝐬𝐬-𝐦𝐨𝐝𝐚𝐥𝐢𝐭𝐲 𝐩𝐚𝐫𝐚𝐦𝐞𝐭𝐫𝐢𝐜 𝐤𝐧𝐨𝐰𝐥𝐞𝐝𝐠𝐞
