Ronghang Hu
29 posts
@RonghangHu
multimodal at @xAI | previously at Meta FAIR
Introducing Quality mode on Grok Imagine – powered by our most advanced image generation model. Quality mode gives you enhanced details, stronger text rendering, and higher levels of creative control. Now available on web and mobile. Try it at grok.com/imagine
We’re releasing SAM 3.1: a drop-in update to SAM 3 that introduces object multiplexing to significantly improve video processing efficiency without sacrificing accuracy. We’re sharing this update with the community to help make high-performance applications feasible on smaller, more accessible hardware. 🔗 Model Checkpoint: go.meta.me/8dd321 🔗 Codebase: go.meta.me/b0a9fb
Grok 4.20 Beta Reasoning makes @xAI a top 5 lab in Vision Arena. Scoring 1240, this model ranks #11 across all Vision models today. Congrats to the @xAI team for this milestone!
Three is a good number. ✅ We have built three massive data centers ✅ Launched a coherent cluster of 330,000 GBs (more to come). ✅ Created over 3,000 jobs. ✅ Paid over $30 million in taxes to local communities. Happy Third Birthday to @xai 🎈
Since xAI was formed just 30 months ago, the small and talented team has made remarkable progress. The future has never looked more exciting!
Collecting a high quality dataset with 4M unique phrases and 52M corresponding object masks helped SAM 3 achieve 2x the performance of baseline models. Kate, a researcher on SAM 3, explains how the data engine made this leap possible. 🔗 Read the SAM 3 research paper: go.meta.me/6411f7
SAM3 live stream tomorrow; image and video segmentation, text and visual prompts, live coding Nov 20th 4:30 PM CET 10:30 AM ET youtube.com/live/G1AEuFwQr…
We’re on the ground at #ECCV2024 in Milan this week to showcase some of our latest research, new research artifacts and more. Here are 4️⃣ things you won’t want to miss from Meta FAIR, GenAI and Reality Labs Research this week whether you’re here in person or following from your feed. 1️⃣ We’re releasing SAM 2.1 an upgraded version of the Segment Anything Model 2 — and the SAM 2 Developer Suite featuring open source tools for training, inference and demos. New artifacts are live in the repo on GitHub ➡️ go.fb.me/mk6ofh 2️⃣ We’re supporting 10+ presentations and workshops in areas like computer vision for smart glasses and the metaverse, 3D vision for eCommerce, egocentric research with Project Aria and more. 3️⃣ We’re presenting seven orals at ECCV — in addition to the 50+ publications from researchers at Meta that were accepted for this year’s conference. Look out for more details on some of these papers later this week. 4️⃣ Demos and discussions with Meta researchers at our booth all week — come by our booth to discuss projects like SAM 2, Ego-Exo4D, DINOv2 and more.
We’re on the ground at #ECCV2024 in Milan this week to showcase some of our latest research, new research artifacts and more. Here are 4️⃣ things you won’t want to miss from Meta FAIR, GenAI and Reality Labs Research this week whether you’re here in person or following from your feed. 1️⃣ We’re releasing SAM 2.1 an upgraded version of the Segment Anything Model 2 — and the SAM 2 Developer Suite featuring open source tools for training, inference and demos. New artifacts are live in the repo on GitHub ➡️ go.fb.me/mk6ofh 2️⃣ We’re supporting 10+ presentations and workshops in areas like computer vision for smart glasses and the metaverse, 3D vision for eCommerce, egocentric research with Project Aria and more. 3️⃣ We’re presenting seven orals at ECCV — in addition to the 50+ publications from researchers at Meta that were accepted for this year’s conference. Look out for more details on some of these papers later this week. 4️⃣ Demos and discussions with Meta researchers at our booth all week — come by our booth to discuss projects like SAM 2, Ego-Exo4D, DINOv2 and more.
Introducing Meta Segment Anything Model 2 (SAM 2) — the first unified model for real-time, promptable object segmentation in images & videos. SAM 2 is available today under Apache 2.0 so that anyone can use it to build their own experiences Details ➡️ go.fb.me/p749s5
Excited to share Worldsheet, a method to synthesize novel views with large camera changes from a *single* image. Turns out, simply shrink-wrapping a mesh sheet onto the image captures 3D well enough to render photorealistic far-away views. w/ @RonghangHu worldsheet.github.io
Looking forward to the lineup of talks, papers, and panel discussions at tomorrow (Friday)'s ALVR workshop on language grounding (images, video, embodied control...): Program: #program" target="_blank" rel="nofollow noopener">alvr-workshop.github.io/#program
Papers: #accepted-papers" target="_blank" rel="nofollow noopener">alvr-workshop.github.io/#accepted-pape… Zoom link: underline.io/events/122/ses… #NAACL2021
