Behzad Bozorgtabar

Computer Vision Group Leader & Lecturer at EPFL
Senior Scientist at CHUV-EPFL

email: behzad.bozorgtabar [at] epfl [dot] ch

RESEARCH UPDATES

Our paper, “A Simple Framework for Open-Vocabulary Zero-Shot Segmentation,” has been accepted at ICLR 2025! 🚀

About Me

My principal research area lies at the intersection of computer vision and medical image analysis and machine learning. I specialize in self-supervised learning, test-time adaptation techniques, and multimodal foundation models, aiming to create robust AI systems capable of adapting to changing environments and data distributions, enhancing decision-making in dynamic settings. My work ultimately strives to bridge the gap between cutting-edge machine learning techniques and their ethical and impactful deployment in real-world applications.

News

EPFL Computer Vision Talks

I am organizing the EPFL Computer Vision Talks
YouTube Channel

Citations

A complete list of my publications and patents can be found at Google Scholar

Recent Selected Publications

*New* ReservoirTTA: Prolonged Test-time Adaptation for Evolving and Recurring Domains
Guillaume Vray^*, Devavrat Tomar^*, Xufeng Gao, Jean-Philippe Thiran, Evan Shelhamer, Behzad Bozorgtabar
arXiv 2025
arXiv

*New* A Simple Framework For Open-Vocabulary Zero-Shot Segmentation
Thomas Stegmüller^*, Tim Lebailly^*, Nikola Dukic, Behzad Bozorgtabar, Tinne Tuytelaars, Jean-Philippe Thiran
ICLR 2025
paper· github (available shortly)

*New* Slide-Level Prompt Learning With Vision Language Models for Few-Shot Multiple Instance Learning In Histopathology
Devavrat Tomar^*, Guillaume Vray^*, Dwarikanath Mahapatra, Sudipta Roy, Jean-Philippe Thiran, Behzad Bozorgtabar
ISBI 2025
paper · github (available shortly)

*New* Self-Supervised Anomaly Segmentation via Diffusion Models with Dynamic Transformer UNet
Komal Kumar, Snehashis Chakraborty, Dwarikanath Mahapatra, Behzad Bozorgtabar, Sudipta Roy
WACV 2025
paper · github

Un-Mixing Test-Time Normalization Statistics: Combatting Label Temporal Correlation
Devavrat Tomar^*, Guillaume Vray^*, Jean-Philippe Thiran, Behzad Bozorgtabar
ICLR 2024
project page · paper · github

CrIBo: Self-Supervised Learning via Cross-Image Object-Level Bootstrapping
Tim Lebailly^*, Thomas Stegmüller^*, Behzad Bozorgtabar, Jean-Philippe Thiran, Tinne Tuytelaars
ICLR 2024 (Spotlight, Top 5%)
project page · paper · github

Distill-SODA: Distilling Self-Supervised Vision Transformer for Source-Free Open-Set Domain Adaptation in Computational Pathology
Guillaume Vray^*, Devavrat Tomar^*, Jean-Philippe Thiran, Behzad Bozorgtabar
IEEE Transactions on Medical Imaging (T-MI) 2024
paper · github

Adaptive Similarity Bootstrapping for Self-Distillation based Representation Learning
Tim Lebailly^*, Thomas Stegmüller^*, Behzad Bozorgtabar, Jean-Philippe Thiran, Tinne Tuytelaars
ICCV 2023
paper · github

CrOC: Cross-View Online Clustering for Dense Visual Representation Learning
Thomas Stegmüller^*, Tim Lebailly^*, Behzad Bozorgtabar, Tinne Tuytelaars, Jean-Philippe Thiran
CVPR 2023
project page · paper · github

TeSLA: Test-Time Self-Learning With Automatic Adversarial Augmentation
Devavrat Tomar, Guillaume Vray, Behzad Bozorgtabar, Jean-Philippe Thiran
CVPR 2023
project page · paper · github

AMAE: Adaptation of Pre-Trained Masked Autoencoder for Dual-Distribution Anomaly Detection in Chest X-Rays
Behzad Bozorgtabar, Dwarikanath Mahapatra, Jean-Philippe Thiran
MICCAI 2023
paper

ScoreNet: Learning Non-Uniform Attention and Augmentation for
Transformer-Based Histopathological Image Classification
Thomas Stegmüller, Behzad Bozorgtabar, Antoine Spahr, Jean-Philippe Thiran
WACV 2023
project page · paper · github

Anomaly Detection and Localization Using Attention-Guided Synthetic Anomaly
and Test-Time Adaptation
Behzad Bozorgtabar, Dwarikanath Mahapatra, Jean-Philippe Thiran
BMVC 2022
paper

OptTTA: Learnable Test-Time Augmentation for
Source-Free Medical Image Segmentation Under Domain Shift
Devavrat Tomar, Guillaume Vray, Jean-Philippe Thiran, Behzad Bozorgtabar
MIDL 2022 (Oral)
project page · paper · github

Self-Rule to Multi-Adapt: Generalized Multi-source Feature Learning
Using Unsupervised Domain Adaptation for Colorectal Cancer Tissue Detection
Christian Abbet, Linda Studer, Andreas Fischer, Heather Dawson, Inti Zlobec,
Behzad Bozorgtabar, Jean-Philippe Thiran
MedIA Journal 2022
paper · github

Self-Supervised Generative Style Transfer for One-Shot Medical Image Segmentation
Devavrat Tomar, Behzad Bozorgtabar, Manana Lortkipanidze, Guillaume Vray,
Mohammad Saeed Rad, Jean-Philippe Thiran
WACV 2022
paper · github

Learning Whole-Slide Segmentation from Inexact and
Incomplete Labels using Tissue Graphs
Valentin Anklin, Pushpak Pati, Guillaume Jaume, Behzad Bozorgtabar,
Antonio Foncubierta-Rodríguez, Jean-Philippe Thiran, Mathilde Sibony,
Maria Gabrani, Orcun Goksel
MICCAI 2021
paper · github

SOoD: Self-Supervised Out-of-Distribution Detection
Under Domain Shift for Multi-Class Colorectal Cancer Tissue Types
Behzad Bozorgtabar, Guillaume Vray, Dwarikanath Mahapatra, Jean-Philippe Thiran
ICCVW 2021
paper · github

Quantifying Explainers of Graph Neural Networks in Computational Pathology
Guillaume Jaume, Pushpak Pati, Behzad Bozorgtabar, Antonio Foncubierta-Rodríguez,
Florinda Feroce, Anna Maria Anniciello, Tilman Rau, Maria Gabrani,
Jean-Philippe Thiran, Orcun Goksel
CVPR 2021
paper · github

Self-Attentive Spatial Adaptive Normalization for Cross-Modality Domain Adaptation
Devavrat Tomar, Manana Lortkipanidze, Guillaume Vray,
Behzad Bozorgtabar, Jean-Philippe Thiran
IEEE Transactions on Medical Imaging (T-MI) 2021
paper · github

Self-Rule to Adapt: Learning Generalized Features from Sparsely-Labeled Data
Using Unsupervised Domain Adaptation for Colorectal Cancer Tissue Phenotyping
Christian Abbet, Linda Studer, Andreas Fischer, Heather Dawson, Inti Zlobec,
Behzad Bozorgtabar, Jean-Philippe Thiran
MIDL 2021
paper · github

Self-Taught Semi-Supervised Anomaly Detection on Upper Limb X-rays
Antoine Spahr , Behzad Bozorgtabar, Jean-Philippe Thiran
ISBI 2021
paper· github

Benefiting from Bicubically Down-Sampled Images for
Learning Real-World Image Super-Resolution
Mohammad Saeed Rad, Thomas Yu, Claudiu Musat, Hazım Kemal Ekenel,
Behzad Bozorgtabar, Jean-Philippe Thiran
WACV 2021
paper

SALAD: Self-Supervised Aggregation Learning for Anomaly Detection on X-Rays
Behzad Bozorgtabar, Dwarikanath Mahapatra, Guillaume Vray, Jean-Philippe Thiran
MICCAI 2020
paper · github

Divide-and-Rule: Self- Supervised Learning for Survival Analysis in Colorectal Cancer
Christian Abbet, Inti Zlobec, Behzad Bozorgtabar, Jean-Philippe Thiran
MICCAI 2020
paper · github

Pathological Retinal Region Segmentation From OCT Images
Using Geometric Relation Based Augmentation
Dwarikanath Mahapatra, Behzad Bozorgtabar, Jean-Philippe Thiran, Ling Shao
CVPR 2020
paper

SynDeMo: Synergistic Deep Feature Alignment for Joint Learning of Depth and Ego-Motion
Behzad Bozorgtabar, Mohammad Saeed Rad, Dwarikanath Mahapatra, Jean-Philippe Thiran
ICCV 2019
paper · supplementary material

SROBB: Targeted Perceptual Loss for Single Image Super-Resolution
Mohammad Saeed Rad, Behzad Bozorgtabar, Urs-Viktor Marti,
Max Basler, Hazım Kemal Ekenel,
Jean-Philippe Thiran
ICCV 2019
paper · supplementary material

Using Photorealistic Face Synthesis and Domain Adaptation
to Improve Facial Expression Analysis
Behzad Bozorgtabar, Mohammad Saeed Rad, Hazım Kemal Ekenel, Jean-Philippe Thiran
FG 2019
paper · github

Learn to Synthesize and Synthesize to Learn
Behzad Bozorgtabar, Mohammad Saeed Rad, Hazım Kemal Ekenel, Jean-Philippe Thiran
CVIU 2019
paper · github

Image-Level Attentional Context Modeling Using Nested-Graph Neural Networks
Guillaume Jaume, Behzad Bozorgtabar, Hazım Kemal Ekenel,
Jean-Philippe Thiran, Maria Gabrani
NeurIPS 2018
paper

MSMCT: Multi-State Multi-Camera Tracker
Behzad Bozorgtabar, Roland Goecke
IEEE TCSVT 2018
paper

Grants

PhD Students

Current Teaching