Nick Stracke
I'm a PhD student in the CompVis group advised by Björn Ommer (LMU). My research focuses on efficient methods to control and alter diffusion T2I models. This includes extending them to perform efficient video generation as well finding disentangled representations for more fine-grained control.
News 🔥
- March 2025 One first author paper and one co-author paper accpeted to CVPR2025.
- July 2024 One first author paper and one co-author paper accpeted to ECCV2024.
- November 2023 Joined the CompVis Group at LMU spearheaded by Björn Ommer.
Research
CleanDIFT: Diffusion Features without Noise
Nick Stracke*, Stefan Andreas Baumann*, Kolja Bauer*, Frank Fundel, Björn Ommer
CVPR 2025
Project Page / arXiv / CodeHow can we get better unsupervised diffusion features? Easy, just remove the noise!
Continuous, Subject-Specific Attribute Control in T2I Models by Identifying Semantic Directions
Stefan Andreas Baumann, Felix Krause, Michael Neumayr, Nick Stracke, Vincent Tao Hu, Björn Ommer
CVPR 2025
Project Page / arXiv / Code / Colab / TwitterT2I diffusion models alredy knew how to do fine-grained control, we just had to learn out how to leverage this capability.
CTRLorALTer: Conditional LoRAdapter for Efficient 0-Shot Control & Altering of T2I Models
Nick Stracke, Stefan A Baumann, Joshua M Susskind, Miguel A Bautista, Björn Ommer
ECCV 2024
Project Page / arXiv / CodeLoRAs don't have to be static! They can also introduce new conditioning into foundation models more efficiently and effectively than previous methods.
Boosting Latent Diffusion with Flow Matching
Johannes Fischer*, Ming Gui*, Pingchuan Ma*, Nick Stracke, Stefan Andreas Baumann, Björn Ommer
ECCV 2024 (Oral)
Project Page / arXiv / CodeMaking high-resolution T2I diffusion fast and increasing resolutions to multiple megapixels by adding flow matching-based superresolution stages in latent space.