Computer Vision and Learning research group at Ludwig Maximilian University of Munich (formerly Computer Vision Group at Heidelberg University)
62
Repositorios públicos
100.472
Total de estrellas
4404
Seguidores
CompVis es un grupo de investigación en Visión por Computadora y Aprendizaje en la Universidad Ludwig Maximilian de Múnich, con una sólida presencia en GitHub. Sus repositorios abarcan una amplia gama de proyectos, destacando herramientas como stable-diffusion y latent-diffusion, desarrollados principalmente en Python y Jupyter Notebook, que reflejan su enfoque en la síntesis de imágenes y modelos generativos.
Principales lenguajes
Repositorios públicos
stable-diffusion
A latent text-to-image diffusion model
latent-diffusion
High-Resolution Image Synthesis with Latent Diffusion Models
taming-transformers
Taming Transformers for High-Resolution Image Synthesis
depth-fm
[AAAI 2025, Oral] DepthFM: Fast Monocular Depth Estimation with Flow Matching
adaptive-style-transfer
source code for the ECCV18 paper A Style-Aware Content Loss for Real-time HD Style Transfer
EDGS
[CVPR 2026] A PyTorch implementation of the paper "EDGS: Eliminating Densification for Efficient Convergence of 3DGS"
vunet
A generative model conditioned on shape and appearance.
geometry-free-view-synthesis
Is a geometric model required to synthesize novel views from a single image?
zigma
A PyTorch implementation of the paper "ZigMa: A DiT-Style Mamba-based Diffusion Model" (ECCV 2024)
metric-learning-divide-and-conquer
Source code for the paper "Divide and Conquer the Embedding Space for Metric Learning", CVPR 2019
fm-boosting
[ECCV 2024, Oral] FMBoost: Boosting Latent Diffusion with Flow Matching
net2net
Network-to-Network Translation with Conditional Invertible Neural Networks
image2video-synthesis-using-cINNs
Implementation of Stochastic Image-to-Video Synthesis using cINNs.
tread
No se proporcionó descripción para este repositorio.
brushstroke-parameterized-style-transfer
TensorFlow implementation of our CVPR 2021 Paper "Rethinking Style Transfer: From Pixels to Parameterized Brushstrokes".
retrieval-augmented-diffusion-models
Official codebase for the Paper “Retrieval-Augmented Diffusion Models”
attribute-control
Fine-Grained Subject-Specific Attribute Expression Control in T2I Models
cleandift
No se proporcionó descripción para este repositorio.
LoRAdapter
No se proporcionó descripción para este repositorio.
imagebart
ImageBART: Bidirectional Context with Multinomial Diffusion for Autoregressive Image Synthesis
discrete-interpolants
The official implementation of "[MASK] is All You Need"
iin
A Disentangling Invertible Interpretation Network
diff2flow
[CVPR 2025] Diff2Flow: Training Flow Matching Models via Diffusion Model Alignment
content-style-disentangled-ST
Content and Style Disentanglement for Artistic Style Transfer [ICCV19]
patch-forcing
[CVPR 2026] Denoising, Fast and Slow: Difficulty-Aware Adaptive Sampling for Image Generation
flow-poke-transformer
No se proporcionó descripción para este repositorio.
RepTok
[ICLR 2026] Adapting Self-Supervised Representations as a Latent Space for Efficient Generation
invariances
Making Sense of CNNs: Interpreting Deep Representations & Their Invariances with Invertible Neural Networks
interactive-image2video-synthesis
No se proporcionó descripción para este repositorio.
unsupervised-disentangling
No se proporcionó descripción para este repositorio.
ipoke
iPOKE: Poking a Still Image for Controlled Stochastic Video Synthesis
SCFlow
[ICCV 2025] SCFlow: Implicitly Learning Style and Content Disentanglement with Flow Models
rayder
No se proporcionó descripción para este repositorio.
long-term-motion
No se proporcionó descripción para este repositorio.
distilldift
[WACV 2025] DistillDIFT: Distillation of Diffusion Features for Semantic Correspondence
unsupervised-part-segmentation
Code for GCPR 2020 Oral : "Unsupervised Part Discovery by Unsupervised Disentanglement"
instant-lora-composition
No se proporcionó descripción para este repositorio.
DisMo
[NeurIPS 2025] DisMo: DIsentangled Motion Representations for Open-World Motion Transfer
maskflow
MaskFlow: Discrete Flows For Flexible and Efficient Long Video Generation
behavior-driven-video-synthesis
No se proporcionó descripción para este repositorio.
DisCLIP
[AAAI 2025] Does VLM Classification Benefit from LLM Description Semantics?
robust-disentangling
Unsupervised Robust Disentangling of Latent Characteristics for Image Synthesis
content-targeted-style-transfer
Content Transformation Block For Image Style Transfer [CVPR19]
metric-learning-divide-and-conquer-improved
Source code for the paper "Improving Deep Metric Learning byDivide and Conquer"
cuneiform-sign-detection-dataset
Dataset provided with the article "Deep learning for cuneiform sign detection with weak supervision using transliteration alignment". It comprises image references, transliterations and sign annotations of clay tablets from the Neo-Assyrian epoch.
visual-search
Visual search interface
cuneiform-sign-detection-code
Code for the article "Deep learning of cuneiform sign detection with weak supervision using transliteration alignment"
hbugen2018
Towards Learning a Realistic Rendering of Human Behavior
magnify-posture-deviations
Unsupervised Magnification of Posture Deviations Across Subjects
weather-rf
[CVPR 2026] Probabilistic Precipitation Nowcasting with Rectified Flow Transformers
Art-fm
[ICCV 2025] Stochastic Interpolants for Revealing Stylistic Flows across the History of Art
myriad
Landing point for "Envisioning the Future, One Step at a Time"
sparse-guidance
No se proporcionó descripción para este repositorio.
wast3d
Official project page for the paper "WaSt-3D: Wasserstein-2 Distance for Scene-to-Scene Stylization on 3D Gaussians"
Characterizing_Generalization_in_DML
No se proporcionó descripción para este repositorio.
AutomaticBehaviorAnalysis_NatureComm
Source Code + Documentation of our Automatic Behavior Analysis Software
cuneiform-sign-detection-webapp
Code for demo web application of the article "Deep learning for cuneiform sign detection with weak supervision using transliteration alignment".
zigma2
No se proporcionó descripción para este repositorio.
network-fusion
No se proporcionó descripción para este repositorio.
cliquecnn
Code for our paper "CliqueCNN: Deep Unsupervised Exemplar Learning" https://arxiv.org/abs/1608.08792
deep_unsupervised_posets
Deep Unsupervised Similarity Learning using Partially Ordered Sets (CVPR17)
mask
The official implementation of "[MASK] is All You Need"
Preguntas frecuentes
¿Qué construye CompVis en GitHub?
CompVis desarrolla una variedad de proyectos relacionados con la visión por computadora, incluyendo modelos de difusión como stable-diffusion y latent-diffusion. Estos proyectos están orientados a la síntesis de imágenes y el aprendizaje profundo, contribuyendo al avance de la investigación en su campo.
¿Qué lenguajes de programación utiliza CompVis?
CompVis utiliza principalmente Python y Jupyter Notebook en sus proyectos, junto con otros lenguajes como JavaScript, HTML, Matlab y CSS. Esta diversidad permite abordar diferentes aspectos de la investigación en visión por computadora y aprendizaje automático.
¿Son públicos los repositorios de CompVis?
Sí, todos los repositorios de CompVis son públicos en GitHub. Esto permite que la comunidad acceda a sus investigaciones y herramientas, promoviendo la colaboración y el intercambio de conocimiento en el ámbito de la visión por computadora.
¿Esta exposición es intencionada?
Monitorea a CompVis - Computer Vision and Learning LMU Munich con RepoGuard y recibe alertas en el momento en que aparece un nuevo repositorio público.
Monitorea esta cuenta