Prof. Dr. Björn Ommer
Leiter, Computer Vision & Learning Group | Ludwig-Maximilians-Universität München
Inhaber, Lehrstuhl Künstliche Intelligenz und Kulturanalytik | Ludwig-Maximilians-Universität München
Publikationen
-
DepthFM: Fast Monocular Depth Estimation with Flow Matching
KonferenzpapierProjekt: KLIMA-MEMES -
Distillation of Diffusion Features for Semantic Correspondence
KonferenzpapierProjekt: KLIMA-MEMES -
Cross-Image-Attention for Conditional Embeddings in Deep Metric Learning
Konferenzpapierhttps://openaccess.thecvf.com/content/CVPR2023/papers/Kotovenko_Cross-Image-Attention_for_Conditional_Embeddings_in_Deep_Metric_Learning_CVPR_2023_paper.pdfProjekt: KLIMA-MEMES -
Diff2Flow: Training Flow Matching Models via Diffusion Model Alignment
Konferenzpapierhttps://doi.org/10.48550/arXiv.2506.02221Projekt: KLIMA-MEMES -
Distillation of Diffusion Features for Semantic Correspondence
DistillDIFT
Konferenzpapierhttps://arxiv.org/abs/2412.03512Projekt: KLIMA-MEMES -
Does VLM Classification Benefit from LLM Description Semantics?
Konferenzpapierhttps://doi.org/10.48550/arXiv.2412.03512Projekt: KLIMA-MEMES -
Fast Monocular Depth Estimation with Flow Matching
DepthFM
Konferenzpapierhttps://doi.org/10.48550/arXiv.2403.13788Projekt: KLIMA-MEMES -
Training Flow Matching Models via Diffusion Model Alignment
Diff2Flow
Konferenzpapierhttps://doi.org/10.48550/arXiv.2506.02221Projekt: KLIMA-MEMES -
Continuous, Subject-Specific Attribute Control in T2I Models by Identifying Semantic Directions
Konferenzpapierhttps://doi.org/10.48550/arXiv.2403.17064Projekt: KLIMA-MEMES -
Diffusion Features without Noise
CleanDIFT
Konferenzpapierhttps://doi.org/10.48550/arXiv.2412.03439Projekt: KLIMA-MEMES -
Token Routing for Efficient Architecture-agnostic Diffusion Training
TREAD
Konferenzpapierhttps://doi.org/10.48550/arXiv.2501.04765Projekt: KLIMA-MEMES -
Understanding Motion Through Sparse Interactions
What If
Konferenzpapierhttps://doi.org/10.48550/arXiv.2510.12777Projekt: KLIMA-MEMES -
Stochastic Interpolants for Revealing Stylistic Flows across the History of Art
Konferenzpapierhttps://openaccess.thecvf.com/content/ICCV2025/html/Ma_Stochastic_Interpolants_for_Revealing_Stylistic_Flows_across_the_History_of_ICCV_2025_paper.htmlProjekt: KLIMA-MEMES -
Implicitly Learning Style and Content Disentanglement with Flow Models
SCFlow
Konferenzpapierhttps://doi.org/10.48550/arXiv.2508.03402Projekt: KLIMA-MEMES -
Governing AI innovation for good
VortragProjekt: KLIMA-MEMES -
Diff2Flow
Training Flow Matching Models via Diffusion Model Alignment
VortragProjekt: KLIMA-MEMES -
Generative KI und die Zukunft der Intelligenz
VortragProjekt: KLIMA-MEMES -
Generative KI und die Zukunft von Intelligenz
VortragProjekt: KLIMA-MEMES


