Log inSign up
Lucas Stoffl
30 posts
@LStoffl

Lucas Stoffl

@LStoffl
Medical AI Postdoc @WeillCornell & @cornell_tech, PhD: @amathislab, Past: @kriegeskortelab, @ctnwaterloo, @apt_csmcr, BSc/MSc @TU_Muenchen
Joined November 2020
495
Following
146
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.
  • @LStoffl
    Lucas Stoffl
    @LStoffl
    Apr 23
    A clinician asks: “Does this biomarker correlate with patient outcome?” Answering it requires segmentation, measurements, code, and careful validation. We built VERITAS 👇🧵 arxiv.org/abs/2604.12144
    Image
    1
  • @LStoffl
    Lucas Stoffl
    @LStoffl
    Oct 1, 2024
    I'm thrilled to present our paper "Elucidating the Hierarchical Nature of Behavior with Masked Autoencoders" at #ECCV24! 🎉 We focus on self-supervised learning for action segmentation to uncover the hierarchical structure of behavior. Paper: ecva.net/papers/eccv_20… 👇🧵
    1
  • @LStoffl
    Lucas Stoffl
    @LStoffl
    Jul 16, 2023
    Delighted that #BUCTD has been accepted at #ICCV! 🙌📚 Grateful for the exceptional teamwork and dedication of @zhoumu53, @TrackingActions, and @TrackingPlumes!
    @TrackingActions
    Mackenzie Weygandt Mathis, PhD
    @TrackingActions
    Jul 14, 2023
    🍾 Super happy to say that #BUCTD was accepted to #ICCV2023! Huge congratulations to @zhoumu53, @LStoffl & @TrackingPlumes, it was so great to work with you all🙏🎉 🥂 🇨🇭@EPFL_en @Brainmind_EPFL @ICepfl @epflSV ⬇️ Details below!
    1
  • @LStoffl
    Lucas Stoffl
    @LStoffl
    Mar 24, 2021
    Super excited to share my first paper as a PhD student!! Big thanks to my wonderful co-authors @vmaxmc2 and @TrackingPlumes! So if you are interested in the first end-to-end trainable method for multi-instance pose estimation, check it out! 😃 arxiv.org/abs/2103.12115
    @amathislab
    A. Mathis Group
    @amathislab
    Mar 24, 2021
    We developed an end-to-end trainable multi-instance pose estimation model with transformers -arxiv.org/abs/2103.12115 - great work by PhD student @LStoffl and Master's student @vmaxmc2!!!
    Image
Advertisement
Advertisement