Generative Vision at Qualcomm AI Research
Amsterdam, Niederlande
Joined April 2017
- Tweets111
- Following405
- Followers323
- Likes121
Jens Petersen retweeted
Excited to share nnInteractive: The nnU-Net moment for 3D Interactive Segmentation! Supports points, scribbles, boxes, & a new lasso prompt for full 3D masks from 2D inputs. Check it out: arxiv.org/abs/2503.08373
github.com/MIC-DKFZ/napari-n…
mitk.org/wiki/MITK-nnInterac…
Jens Petersen retweeted
🌵Presenting our paper on predictive imaging biomarker discovery at #WACV2025 this Sunday!
📄 arxiv.org/abs/2406.02534v2
Thanks to @lukas_kln, @peterjensen_, @VollmuthP, @pfjaeger and @maierhein for working on this together @mic_dkfz!
Join @noor_fathima_ and myself this afternoon at #CVPR2024 to learn more about Clockwork Diffusion :) Highlight poster #345, from 5pm in Arch 4A-E
Interested in fast diffusion models? We will be presenting Clockwork at #CVPR2024.
Paper: arxiv.org/abs/2312.08128
Code: github.com/qualcomm-ai-resea…
Jens Petersen retweeted
The code for clockwork is now open-source!
It allows up to 40% time savings with close to no loss in percept. quality w/o any finetuning required. Add a single line to your pipeline synthesis code and see for yourself!
github.com/Qualcomm-AI-resea…
Do we need to run the whole UNet for all the diffusion steps? No!
Accelerating your Diffusion Model with a simple trick, even without retraining ?!
w/ Amir Ghodrati, @noor_fathima_ , @gsautiere, Fatih Porikli and @peterjensen_
arxiv.org/abs/2312.08128
Code: Stay tuned
Come join our team for an internship 🙆♂️
Multiple PhD internships available in Qualcomm AI Research, Amsterdam
linkedin.com/jobs/view/37982…
Topic:
- Image/video generation and editing
- Efficiency of diffusion models
- Efficient architectures
- Adopting generative models to improve down-stream tasks
DM me for any question!
Jens Petersen retweeted
With a text prompt and a click of a button you can now generate brand new 2D images, 3D images, and even videos. But at what cost? While this technology has great potential, it faces challenges which we will address in this #AI webinar. Sign up here: assets.qualcomm.com/efficien…
Jens Petersen retweeted
nnU-Net has stood the test of time, continues to deliver excellent results and inspires the community as a framework for building new segmentation methods!
It is with great excitement that we announce the release of nnU-Net V2 🎉😍
➡ github.com/MIC-DKFZ/nnUNet
🧵(1/8)
Jens Petersen retweeted
Neural Image Compression with a Diffusion-Based Decoder
abs: arxiv.org/abs/2301.05489
Jens Petersen retweeted
Excited about this #SurgicalDataScience application with potential for relatively fast clinical translation: rdcu.be/cOtQ6! All the details on June 7th at @IPCAI_conf. @uniklinik_hd / @DKFZ project by @haydnspass @doctor_felix (PI) @peterjensen_ @beatpetemueller et al
If you're at #MICCAI2021 this week, check out our poster on "Continuous-Time Deep Glioma Growth Models" (Wed-S2, R071-D). Also interesting if you work with other kinds of stochastic and sparse image time series :) Paper: arxiv.org/abs/2106.12917 Code: github.com/MIC-DKFZ/deep-gli…
@MICCAI_Society @maierhein @FabianIsensee1 @GregorKoe @pfjaeger @d_zimmerer @wick_wolfgang @VollmuthP
Jens Petersen retweeted
🙌🏻 Polarizing or already arrived in everyday life?
👋 "Ethics and AI: How to successfully integrate both into businesses" - on July 27th at 4 – 6 pm.
Interactive talk about, #privacy and #bias in algorithms and 1:1 sessions.
📢 So join us now: lnkd.in/gNfbvHq
It's been a while, but the #heidelbergai YT channel has 3 new videos on graph partitioning (@FredHamprecht ), importance of expert knowledge in MIA (@lena_maierhein ), large language models (@TimoDenk ): youtube.com/channel/UCfHWBne… . w/ @pfjaeger @GregorKoe @maierhein
Jens Petersen retweeted
After two years of work, the preprint of largest ever medical image segmentation challenge - the Medical Segmentation Decathlon - is out: arxiv.org/abs/2106.05735. It's incredible how far we've come in making image segmentation a commodity tool. More here medicaldecathlon.com
Jens Petersen retweeted
From Science to Startup to Siemens 👩🔬🛰️
...How? Join the innovation journey of AI on July 1st, 4 – 6 pm!
An interactively designed panel discussion and personal 1:1 session with the speakers and organizers await you.
📢Register here: lnkd.in/dqGwxJi
Jens Petersen retweeted
Limitations of image processing metrics at a glance: arxiv.org/pdf/2104.05642.pdf. Feel free to contribute to the next version of this dynamic article! @IPCAI_conf @MiccaiStudents @WomenInMICCAI @HIDAdigital @helmholtz_ai. Current version compiled by authors from 35 labs:
Jens Petersen retweeted
Looking forward to the @DataScienceDKFZ and heidelberg.ai seminar talk by @FredHamprecht tomorrow. Talk will be available at dkfz.de/en/datascience/semin…. You can register at heidelberg.ai/ @mic_dkfz @DKFZ_CAMI_lab @peterjensen_ @pfjaeger @doreen_HeNoe
Jens Petersen retweeted
BVM 2021 is happy to announce four virtual tutorials at 07.03.2021
Tutorial 1: Advanced #DeepLearning by @mic_dkfz @peterjensen_, @maierhein et al.
Registration in combination with BVM conference or separately:
bvm-workshop.org/registrieru…
@ReMIC_OTH
@BVM_Community
@DGBMT
Jens Petersen retweeted
We're now accepting applications for our school's third year!
13 fully funded PhD positions @KITKarlsruhe @DKFZ @UniHeidelberg for candidates who want to be the next leaders in #AI and #DataScience for healthcare. Application deadline March 1!
Jens Petersen retweeted
Exciting opportunity: The newly founded Applied Computer Vision Lab, headed by @FabianIsensee1, is looking for three excellent machine learning engineers with a background in deep learning and computer vision. Apply here until Jan 10th 2021: tiny.cc/z5o7tz. Please share!