Log inSign up
Davis Rempe
199 posts
@davrempe

Davis Rempe

@davrempe
Research Scientist @ NVIDIA working on human motion generation and perception. Previously PhD at Stanford.
San Francisco, CA
davrempe.github.io
Joined April 2012
570
Following
1,148
Followers
RepliesRepliesRepostsRepostsMediaMedia

Log in or sign up for X

See what’s happening and join the conversation

Continue with phone
or
Log in with username or email
Terms·Privacy·Cookies·Accessibility·Ads Info·© 2026 X Corp.
  • @davrempe
    Davis Rempe
    @davrempe
    Jul 20
    I am at SIGGRAPH this week! ARDY and MotionBricks will be presented in the Motion Generation, Warping & Control session on Tuesday at 2pm (Room 403A). Come to learn about real-time motion generation and chat with the team.
    @kaifengzhao
    Kaifeng Zhao
    @kaifengzhao
    Jul 10
    Introducing ARDY — real-time, controllable 3D human motion generation Streaming text prompts + flexible kinematic constraints — root waypoints, full-body keyframes, joint targets — all at interactive speed To be presented at SIGGRAPH 2026, Paper & demo research.nvidia.com/labs/sil/proje… 1/5
    Image
    00:00
    2
  • @davrempe
    Davis Rempe
    @davrempe
    Jul 15
    ARDY now has a live demo online thanks to the folks @HuggingApps! Try it out for yourself.
    @HuggingApps
    Hugging Apps
    @HuggingApps
    Jul 15
    NVIDIA’s ARDY is a glimpse of where AI animation is heading: real time, open source and you can play with it right now on @huggingface Spaces Type what the character should do and generate a motion sequence in seconds. Give a sentence a body. huggingface.co/spaces/hugging…
    Image
    00:00
  • @davrempe
    Davis Rempe
    @davrempe
    Jul 13
    Our new method ARDY keeps the text and pose controllability that we get with Kimodo, but runs autoressively in real time! This unlocks human motion generation for a variety of interactive applications. Come chat with us at SIGGRAPH next week to learn more.
    @kaifengzhao
    Kaifeng Zhao
    @kaifengzhao
    Jul 10
    Introducing ARDY — real-time, controllable 3D human motion generation Streaming text prompts + flexible kinematic constraints — root waypoints, full-body keyframes, joint targets — all at interactive speed To be presented at SIGGRAPH 2026, Paper & demo research.nvidia.com/labs/sil/proje… 1/5
    Image
    00:00
  • @davrempe
    Davis Rempe
    @davrempe
    Apr 29
    Kimodo is good for motion authoring, but what about real-time applications? Enter MotionBricks! At its core, MotionBricks is a high-quality motion in-betweener. When combined with smart keyframing, it produces amazing locomotion and environment interactions, all in real time.
    @TingwuWang
    tingwu.wang
    @TingwuWang
    Apr 29
    What is missing to bring real-time motion research into AAA games and real-world robotics? We present MotionBricks, a step toward bridging this gap with two key components: - a single generative latent motion backbone covering 350,000+ motion skills, running at 15,000 FPS with 2
    Image
    00:00
    1
  • @davrempe
    Davis Rempe
    @davrempe
    Apr 10
    Are you training motion gen models on @TheBonesStudio’s SEED dataset? Evaluate and compare your models on our new Kimodo Motion Generation Benchmark! It includes an extensive text-to-motion test suite and code for computing standardized metrics. github.com/nv-tlabs/kimod…
    @davrempe
    Davis Rempe
    @davrempe
    Mar 17
    Need high-quality motion for humanoid robots or digital humans? Meet Kimodo: our new diffusion model trained on 700 hours of optical mocap data for easy, controllable, and high-fidelity motion generation. @NVIDIAAI research.nvidia.com/labs/sil/proje…
    Image
    00:00
    1
Advertisement
Advertisement