Log inSign up
user avatar

Jay Karhade

@JayKarhade
PhD Robotics @CMU_Robotics, Computer Vision, Robotics.
jaykarhade.github.io
Joined September 2022
465
Following
626
Followers
RepliesMedia
  • Pinned
    user avatar
    Jay Karhade
    @JayKarhade
    Dec 12, 2025
    Introducing Any4D, a unified transformer for fully feed-forward, dense, metric-scale 4D reconstruction from flexible inputs! Any4D regresses per-pixel motion + geometry across frames in one pass — 15× faster, 2–3× more accurate reconstructions ⚡📈 Details + code below 👇
    Image
    00:00
  • user avatar
    Jay Karhade
    @JayKarhade
    Jul 18
    Very cool work! Totally expect video models to be a good prior for predicting 3D flow!
    user avatar
    Homanga Bharadhwaj
    @mangahomanga
    Jul 17
    Introducing MotionForesight: A very simple approach re-purposing video models for forecasting future 3D scene flow given a short observed video context Trained with just 40k RGB human videos, it generalizes well to completely new home/office scenes! 1/n
    Image
    00:00
  • user avatar
    Jay Karhade
    @JayKarhade
    Jun 23
    Really cool work, one of the more impressive visuals I have seen for 4D reconstruction!
    user avatar
    Yehonathan Litman
    @yehonation
    Jun 23
    To be honest, training on handmade 4D asset datasets is a dead-end. Almost all 4D asset data is synthetic and diverse real data barely exists, so models trained on it struggle to reconstruct objects that deform, get occluded, and move freely about the scene. Our new work,
    Image
    00:00
  • user avatar
    Jay Karhade
    @JayKarhade
    Jun 6
    Well deserved! D4RT has put 4D on the center stage! If you'd like to chat about the future of 4D, swing by tomorrow at the Any4D poster at ExHall F32 11:45!
    user avatar
    #CVPR2026
    @CVPR
    Jun 5
    Announcing the official #CVPR2026 Best Paper Award Winner! 🏆 Congratulations to the authors for their landmark contributions to the field!👏👏
    Image