🐝 Explore HIVE: Pioneering RLHF for text-to-image generation & editing. Tailored for diffusion models, no extra compute costs. Presented at #CVPR24. #RLHF #ComputerVision 🐝 bit.ly/3MbcYZ7

💥 Motion capture, perfected 💥 Our new method learns compact temporal trajectories for Gaussian splatting, 🧑‍💻 enabling efficient representation of dynamic scenes. 🔥🚀 #CVPR24 #computergraphics #3D

② 続くIT-02 “Active Domain Adaptation with False Negative Prediction for Object Detection”も #CVPR24 採択テーマ！AIモデルの見落としに着目したアクティブドメイン適応技術 (中部大 山下先生 Takayoshi Yamashita 連携) #MIRU2024 IT-01,02は8月7日(水) IS-1Bでポスターも📄 news.panasonic.com/jp/press/jn240…

#MIRU2024 (8/6-9) で発表予定の6件をご紹介✨ ① 8/6(火) 招待講演 IT-01“Deep Single Image Camera Calibration by Heatmap Regression to Recover Fisheye Images Under Manhattan World Assumption”は #CVPR24 採択のカメラ校正技術 (中部大 山下先生 Takayoshi Yamashita 連携) news.panasonic.com/jp/press/jn240…

Coming to Ecological Society in Long Beach next week? After the massive success of the AI for Conservation meetup at #CVPR24 we're doing it again :) ESA Conservation Tech Meetup at 7pm Monday August 5th, at Altar Society Brewing! See you there fellow geeks 😜 maps.app.goo.gl/QwBKQPofRKQJbG…

#CVPR24, #ECCV2024, #ICCV, #NEURIPS gibi konferanslarla ilgilenenleri, #yapayGörme, #örüntüTanıma, #yapayÖğrenme vs. konularda çalışanları, 2016'da kurduğumuz iletişim listemize bekleriz: groups.google.com/g/cvpr-news-tr/

#TBT to last week, where we proudly presented HIVE, our pioneering project using human feedback (RLHF) for text-to-image generation and editing. Thanks #CVPR24 for another great year! 🎓blog.salesforceairesearch.com/hive/ #ComputerVision #CVPR #HIVE

Here are the slides from my #CVPR24 tutorial on Evaluation of video generative models tinyurl.com/mtvyycen I'd love to hear any feedback on the material 🙏

Finished up #CVPR24 with a field visit to two in-river SONAR deployments used to monitor salmonids by Washington Department of Fish & Wildlife and NOAA as part of a workshop bringing together agencies and NGOs using this technology to improve the sustainability of fisheries from California up to Alaska 🐟🐟🐟

My main insight from #CVPR24 is if you think data driven learning in simulators and/or in the wild will solve most AI problems, buy $NVDA. Otherwise don't.

To vote, you need your #CVPR24 registration number - find it in your registration email. Also make sure to vote No ❌ for “Motion 2: Text-Only Rebuttals.” I use figures often as concise additional clarifications - figures are not only for additional experiments. CVPR != SIGGRAPH

OmniGlue on display at #CVPR24 Thanks for presenting our work! Hanwen Jiang André Araujo Full paper: hwjiang1510.github.io/OmniGlue/

Highlight from #CVPR24 During the Day 3 poster session, I stopped by to have a chat with Prof. Michael Black of the Intelligent Systems and Chief Scientist at Meshcapade. It was nothing short of interesting conversation.

#CVPR2024 2024 last day! Wanna know how to train a segmentation model with only text labels? Stop by our CoDe poster😎 [⏰Time] June 21 (Fri.) 5:00 pm - 6:30 pm [📍Location] Poster Session 6, #255, Exhibit Hall (Arch 4A-E) [Website] 072jiajia.github.io/image-text-co-… #cvpr #cvpr2024 #cvpr24

Couldn’t agree more! I couldn’t talk/meet with the authors that are super relevant to my work. At such a large scale conference, if you don’t go to their poster session you will never catch the people that you don’t know them before! (Unless you are a senior ;( #CVPR24 #CVPR2024

Attended a great #CVPR24 in Seattle. Most important news of the week for me is that Uber costs me as much as plane ticket from SEA to SFO. Can't wait for robotaxi.

🚨 Happening tomorrow afternoon at poster 102 #CVPR24 🚨 If you are using denoising architectures, come hear about a cool trick that boosts results and that has fun mathematical implications 🥳💫🔥 Code: github.com/matthieutrs/Eq… Camera ready: bit.ly/4cqqvXY