Timo Kepp

Zeit, neue Wege zu gehen: Wir beenden unsere Aktivitäten auf X. Bleiben Sie mit uns im Dialog – auf unseren anderen Kanälen. uni-luebeck.de/aktuelles/nach… #eXit #WissXit #ByeByeX #QuitX

Huge congratulations to Joshua Niemeijer for winning best paper award! Thanks to everyone for coming and supporting each other today! Let’s see each other again in Daejeon! #MICCAI2024

Today, I’m excited to share with you all the fruit of our effort at @OpenAI to create AI models capable of truly general reasoning: OpenAI's new o1 model series! (aka 🍓) Let me explain 🧵 1/

We will be presenting 2 studies at this year's @EURETINA conference in Barcelona! #EURETINA2024 #OphthalmologyResearch 1/7👇

Exciting news! We are releasing another major update to TotalSegmentator: It can now segment 16 different brain structures in CT images. Try out the new "brain_structures" task.

My paper "Training-free Graph Neural Networks and the Power of Labels as Features" has been accepted to #TMLR 🎉 I proposed training-free (and optionally trained) GNNs. Paper📜：arxiv.org/abs/2404.19288 Code📁：github.com/joisino/laf

🆕 Research paper from GenAI at Meta: Imagine yourself: Tuning-Free Personalized Image Generation. Research paper ➡️ go.fb.me/wre8f0 Want to try it? The feature is available now as a beta in Meta AI for users in the US.

Early Access Paper Just Posted: A Novel Poroelastography Method for High-quality Estimation of Lateral Strain, Solid Stress and Fluid Pressure In Vivo. Read the paper: ieeexplore.ieee.org/document/10623…

Flux.1-DEV with ControlNet Canny!🔥🔥

Hidden deep inside the depth of internet, a treasure. #probability

Weakly supervised ROI proposal networks (WSRPN) provide a fully differentiable way to train object detection models with image-level labels only. They significantly improve over the previous state-of-the-art in pathology detection on chest radiographs. ieeexplore.ieee.org/document/10613…

arXiv -> alphaXiv Students at Stanford have built alphaXiv, an open discussion forum for arXiv papers. @askalphaxiv You can post questions and comments directly on top of any arXiv paper by changing arXiv to alphaXiv in any URL!

With larger and larger diffusion transformers coming up, it's becoming increasingly important to have some good quantization tools for them. We present our findings from a series of experiments on quantizing different diffusion pipelines based on diffusion transformers. We demonstrate excellent memory savings with a bit of sacrifice on inference latency which is expected to improve in the coming days. Diffusers 🤝 Quanto ❣️ 🧵 1/7

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

Meta Segment Anything Model v2 (SAM 2) is out. Can segment images and videos. Open source under Apache-2 license. Web demo, paper, and datasets available. Amazing performance.

@jeremyphoward reached out to me yesterday asking me if I wanted to get a sneak peek at a new library he was working on. I liked the experience so much that I ended up making an intro tutorial for it ... as well as a plugin 😅 youtube.com/watch?v=4En57Z…

“We fine tune the last layer”

🚀🩻 Great News! Our paper on Resource-efficient Forward-Forward networks without Back-propagation for Medical Imaging Analysis has been accepted to the #MLMI workshop at MICCAI 2024!🌍💫 @MICCAI_Society @BorderlessSci @aibe_fau @UniFAU @BernhardKainz1 arxiv.org/abs/2406.14038

Our own @hannah_eichhorn on PHIMO: Physics-Informed Motion Correction of GRE MRI for T2* Quantification 👏👏👏 @HelmholtzIml @NMRMgroup (#ChristinePreibisch)

In an upcoming MICCAI paper, we present a latent variable model designed to model complex heart structures and to precisely capture the intricate relationships between heart components. infoscience.epfl.ch/handle/20.500.… #deeplearning #MedicalImaging