Log inSign up
Nikita Karaev
116 posts
user avatar
Nikita Karaev
@n_karaev
Building general purpose robotics at Amazon FAR. @a16z Speedrun Scout. Previously founder @ pixelwise.ai / @AIatMeta / @Oxford_VGG PhD. nikitakaraevv.github.io
London, England
linkedin.com/in/nikitakarae…
Joined October 2013
559
Following
3,257
Followers

New to X?

Sign up now to get your own personalized timeline!

Create account

By signing up, you agree to the Terms of Service and Privacy Policy, including Cookie Use.

Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
Don't miss what's happening
People on X are the first to know.
Log inSign up
  • Pinned
    user avatar
    Nikita Karaev
    @n_karaev
    Oct 6, 2025
    🚨Big news - our team is joining the Frontier AI and Robotics (FAR) lab at @amazon to keep building the future of robotics. Excited to be joining such an inspiring team and to work with @peterxichen, @pabbeel, @rocky_duan, @akanazawa and many others at FAR.
    82K
  • user avatar
    Nikita Karaev
    @n_karaev
    Sep 10, 2025
    Personal update - I'm now officially an a16z @speedrun scout. This means I can write you a 10k$ check within 24 hours. If you're an early-stage founder or thinking about starting a company and have done something cool before - my DMs are open! I'm mostly into AI/robotics, but
    226K
  • user avatar
    Nikita Karaev
    @n_karaev
    Jan 3, 2024
    1/5 Last week, we released an updated version of CoTracker, our transformer-based model for tracking any point in a video. It’s now capable of tracking 10x more points than before!
    00:00
    64K
  • user avatar
    Nikita Karaev
    @n_karaev
    Oct 1, 2024
    Tomorrow at #ECCV2024 I’m presenting CoTracker: It is better to track together! We jointly track thousand of points in videos through occlusions. Come see our poster! 🖼️ Poster #253 @ 10:30 on Wednesday Project page: co-tracker.github.io
    00:00
    17K
  • user avatar
    Nikita Karaev
    @n_karaev
    Feb 16, 2024
    CoTracker tracks for this SORA video. 3D geometry is just 🤯
    00:00
    user avatar
    Ryan Hickman
    @ryanmhickman
    Feb 16, 2024
    Replying to @AjdDavison
    I wanna see @n_karaev run CoTracker on these to highlight keypoints moving in the scene!
    41K
  • user avatar
    Nikita Karaev
    @n_karaev
    Sep 11, 2025
    Wow I didn't expect to receive so many DMs (400+), I'll go through them one by one. The pattern I noticed so far, which wasn't obvious to me a year ago - most people jump straight into pitching their idea without saying a word about the team. The team is actually *the* most
    19K
  • user avatar
    Nikita Karaev
    @n_karaev
    Jul 8, 2025
    ⚡️Today we’re releasing SpatialTrackerV2, the first feedforward model for dynamic 3D reconstruction and point tracking in the world space. Project page: spatialtracker.github.io 🤗 demo: huggingface.co/spaces/Yuxihen…
    00:00
    20K
  • user avatar
    Nikita Karaev
    @n_karaev
    May 4, 2023
    My first PhD paper with @Oxford_VGG and @metaai, on depth estimation from dynamic stereo videos has been accepted at CVPR 2023!! ☺️ #CVPR2023 #cvpr #computervision #3D
    user avatar
    AK
    @_akhaliq
    May 4, 2023
    DynamicStereo: Consistent Dynamic Depth from Stereo Videos abs: arxiv.org/abs/2305.02296 project page: dynamic-stereo.github.io
    00:00
    34K
  • user avatar
    Nikita Karaev
    @n_karaev
    Feb 28, 2025
    Microsoft @jw2yang4ai just released a VLM, Magma, where they presented a scalable way to train policy models on real data by annotating 10M videos using CoTracker, point tracking model that we developed while back at Meta AI. Magma: microsoft.github.io/Magma/
    00:00
    7.9K
  • user avatar
    Nikita Karaev
    @n_karaev
    Mar 17, 2025
    Check out our latest work, VGGT! ⚡️ It’s a fast transformer that predicts cameras, point maps, depth and 3D point tracks. VGGT + CoTracker significantly outperforms CoTracker in 2D point tracking.
    user avatar
    Jianyuan
    @jianyuan_wang
    Mar 17, 2025
    Introducing VGGT (CVPR'25), a feedforward Transformer that directly infers all key 3D attributes from one, a few, or hundreds of images, in seconds! No expensive optimization needed, yet delivers SOTA results for: ✅ Camera Pose Estimation ✅ Multi-view Depth Estimation ✅ Dense
    00:00
    14K
  • user avatar
    Nikita Karaev
    @n_karaev
    Jan 3, 2024
    Replying to @n_karaev
    5/5 You can try our 🤗 HF demo here: huggingface.co/spaces/faceboo… And check out the updated paper: arxiv.org/abs/2307.07635
    CoTracker - a Hugging Face Space by facebook
    From huggingface.co
    11K
  • user avatar
    Nikita Karaev
    @n_karaev
    Jun 21, 2023
    Today at #CVPR2023 I am presenting our paper "DynamicStereo: Consistent Dynamic Depth from Stereo Videos" with @MetaAI and @Oxford_VGG. Come visit our Poster #83 with an AR demo in Meta Quest Pro at 4:30pm! Website➡️dynamic-stereo.github.io Arxiv ➡️ arxiv.org/abs/2305.02296
    GIF
    3.4K
  • user avatar
    Nikita Karaev
    @n_karaev
    Jul 18, 2023
    Today we released CoTracker, a model that bridges the gap between long-term point tracking and optical flow. Please try our Colab demo and explore more results on the project page at co-tracker.github.io!
    user avatar
    AK
    @_akhaliq
    Jul 18, 2023
    CoTracker: It is Better to Track Together paper page: huggingface.co/papers/2307.07… Methods for video motion prediction either estimate jointly the instantaneous motion of all points in a given video frame using optical flow or independently track the motion of individual points
    00:00
    5.7K
  • user avatar
    Nikita Karaev
    @n_karaev
    Sep 11, 2025
    Replying to @n_karaev
    I see some questions about the check size. A $10k check is indeed quite symbolic, Speedrun itself invests much more. The main value lies in the connections, experience and the support for applying to speedrun, not in the money itself.
    7.4K