Derui Zhu; Dingfan Chen; Jens Grossklags; Lei Ma; Mario Fritz · cispa.saarland

Our group is working on Trustworthy AI and AI Security and Safety in various domains like Cybersecurity, Health, Computer Vision, Natural Language Processing.

We are looking for PhD students and Post-Docs! Please get in touch.
We also offer student projects, bachelor/master thesis, internships.

Recent projects and initiatives related to trustworthy AI/ML, health, privacy:

Recent activities w.r.t. policy, regulation, outreach

  • Frontier AI expert Findings on EU competitiveness, sovereignty and security
  • Member of the AI Act Scientific Panel
  • Expert for AI Cybersecurity Uplift Study “Toward Quantitative Modeling of Cybersecurity Risks Due to AI Misuse”
  • Independent Expert study on technical means for implementing article 50 of EU AI Act towards the code of practice
  • Scientific Lead of the CISPA-ELLIS Summerschool 2025 on “Trustworthy AI – Secure and Safe Foundation Models”
  • Panel on “Loss of Control” at EU AI Office Workshop on “Towards defining best practices for systemic risk evaluations”
  • Keynote on “Security and Safety of LLMs and AI Agents” at 18th ACM International Conference on Web Search and Data Mining
  • Talk on “Fundamental Risks in the Current Deployment of General-Purpose AI Models: What Have We (Not) Learnt From Cybersecurity?” at European AI Office workshop “Evaluating General-Purpose AI Models with Systemic Risk”
  • Expert opinion for evidence review report “Successful and timely uptake of artificial intelligence in science in the EU”
  • Co-Author “ELSA Strategic Research Agenda: Facing the Grand Challenges of Secure and Safe AI”
  • Co-Author “AI, Data, and Robotics ‘made in Europe’: Research agendas from the EU AI and robotics Networks of Excellence”
  • Lecture at Artificial Intelligence Doctoral Academy (AIDA) on “Trustworthy AI and A Cybersecurity Perspective on Large Language Models”

2026

Journal Articles

IPAuditor: Privacy Violation in Data Release Using Diffusion-based Generative Models

IPAuditor: Privacy Violation in Data Release Using Diffusion-based Generative Models Journal Article

In: Transactions on Dependable and Secure Computing, 2026.

Functional rescue and AI analysis of a human inactivating GPCR mutation using a small molecule

The science and practice of proportionality in AI risk evaluations

Carlos Mougan; Lauritz Morlock; Jair Aguirre; James R. M. Black; Jan Brauner; Simeon Campos; Sunishchal Dev; David Fernández Llorca; Alberto Franzin; Mario Fritz; Emilia Gómez; Friederike Grosse-Holz; Eloise Hamilton; Max Hasin; Jose Hernandez-Orallo; Dan Lahav; Luca Massarelli; Vasilios Mavroudis; Malcolm Murray; Patricia Paskov; Jaime Raldua; Wout Schellaert

The science and practice of proportionality in AI risk evaluations Journal Article

In: Science, vol. 391, no. 6787, pp. 769-771, 2026.

An exploratory survey study toward a conceptual framework for structuring the integration of privacy-enhancing technologies in federated learning and analytics

Towards Useful and Private Synthetic Omics: Community Benchmarking of Generative Models for Transcriptomics Data

How foundation models will revolutionize robot swarms

What Survives Privatization? A Guide to Structure and Utility in Differentially Private Genome-Wide Association Studies

Proceedings Articles

The Oracle's Gambit: A Game-Theoretic Framework for Responsible AI Release

MATRA: Modeling the Attack Surface of Agentic AI Systems - OpenClaw Case Study

Certified Circuits: Stability Guarantees for Mechanistic Circuits

Position: Trustworthy AI Suffers from Invariance Conflicts and Causality is The Solution

Position: Safety Must Precede the Deployment of Open-Ended AI Agents

Justice in Judgment: Unveiling (Hidden) Bias in LLM-assisted Peer Reviews

ProxyPrompt: Securing System Prompts against Prompt Extraction Attacks

Enhancing Cyber Attack Autonomy Through Multi-Agent Reinforcement Learning (MARL)

Christoph R. Landolt; Julian Jang-Jaccard; Valentin Mulder; Roland Meier; Christoph Würsch; Mario Fritz

Enhancing Cyber Attack Autonomy Through Multi-Agent Reinforcement Learning (MARL) Proceedings Article

In: 18th International Conference on Cyber Conflict: Securing tomorrow (CyCon), 2026.

Inspectable AI for Science: A Research Object Approach to Generative AI Governance

Ruta Binkyte; Sharif Abuadbba; Chamikara Mahawaga Arachchige; Ming Ding; Natasha Fernandes; Mario Fritz

Inspectable AI for Science: A Research Object Approach to Generative AI Governance Proceedings Article

In: 1st Workshop on Metascience and Critical Reflections in Security & Privacy at S&P (S&P-W), 2026.

Funny or Persuasive, but Not Both: Evaluating Fine-Grained Multi-Concept Control in LLMs

Deepfake Detection that Generalizes Across Benchmarks

IV Co-Scientist: Multi-Agent LLM Framework for Instrumental Variable Discovery

Technical Reports

Scalable Delphi: Large Language Models for Structured Risk Estimation

Certified Circuits: Stability Guarantees for Mechanistic Circuits

2025

Journal Articles

NeurIPS 2023 Competition: Privacy Preserving Federated Learning Document VQA

Marlon Tobaben; Mohamed Ali Souibgui; Rubèn Tito; Khanh Nguyen; Raouf Kerkouche; Kangsoo Jung; Joonas Jälkö; Lei Kang; Andrey Barsky; Vincent Poulain D'Andecy; Aurélie Joseph; Aashiq Muhamed; Kevin Kuo; Virginia Smith; Yusuke Yamasaki; Takumi Fukami; Kenta Niwa; Iifan Tyou; Hiro Ishii; Rio Yokota; Ragul N; Rintu Kutum; Josep Lladós; Ernest Valveny; Antti Honkela; Mario Fritz; Dimosthenis Karatzas

NeurIPS 2023 Competition: Privacy Preserving Federated Learning Document VQA Journal Article

In: Transactions on Machine Learning Research (TMLR), 2025.

DP-2Stage: Adapting Language Models as Differentially Private Tabular Data Generators

Taxonomy, Opportunities, and Challenges of Representation Engineering for Large Language Models

Proceedings Articles

Justice in Judgment: Unveiling (Hidden) Bias in LLM-Assisted Peer Reviews

Beyond Steering: Evaluating Fine-Grained and Multi-Concept Control in LLMs

Arya Labroo; Ivaxi Sheth; Vyas Raina; Amaani Ahmed; Mario Fritz

Beyond Steering: Evaluating Fine-Grained and Multi-Concept Control in LLMs Proceedings Article

In: NeurIPS Workshop on Evaluating the Evolving LLM Lifecycle: Benchmarks, Emergent Abilities, and Scaling, 2025.

MIBP-Cert: Certified Training against Data Perturbations with Mixed Integer Bilinear Programs

MaxSup: Overcoming Representation Collapse in Label Smoothing

Can LLMs Propose Instrumental Variables for Causal Reasoning?

Ivaxi Sheth; Zhijing Jin; Bryan Wilder; Dominik Janzing; Mario Fritz

Can LLMs Propose Instrumental Variables for Causal Reasoning? Proceedings Article

In: NeurIPS 2025 Workshop on CauScien: Uncovering Causality in Science, 2025.

Context-Aware Reasoning On Parametric Knowledge for Inferring Causal Variables

Accept or Deny? Evaluating LLM Fairness and Performance in Loan Approval across Table-to-Text Serialization Approaches

Automated Detection of Abnormalities in Zebrafish Development

Sarath Sivaprasad; Hui-Po Wang; Anna-lisa Jäckel; Jonas Baumann; Carole Baumann; Jennifer Herrmann; Mario Fritz

Automated Detection of Abnormalities in Zebrafish Development Proceedings Article

In: 28th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI), 2025.

Safety is Essential for Responsible Open-Ended Systems

Causality Is Key to Understand and Balance Multiple Goals in Trustworthy ML and Foundation Models

CausalGraph2LLM: Evaluating LLMs for Causal Queries

Balancing Diversity and Risk in LLM Sampling: How to Select Your Method and Parameter for Open-Ended Text Generation

A Theory of LLM Sampling: Part Descriptive and Part Prescriptive

EARL: Early Intent Recognition in GUI Tasks Using Theory of Mind

Shraddha Vijay Pawar; Balavarun Pedapudi; Pramod Kaushik; Sarath Sivaprasad; Mario Fritz; Shirish Karande

EARL: Early Intent Recognition in GUI Tasks Using Theory of Mind Proceedings Article

In: ICML 2025 Workshop on Computer Use Agents (ICML-W), 2025.

Stealix: Model Stealing via Prompt Evolution

Zhixiong Zhuang; Hui-Po Wang; Maria-Irina Nicolae; Mario Fritz

Stealix: Model Stealing via Prompt Evolution Proceedings Article

In: International Conference on Machine Learning (ICML), 2025.

Pixel-level Certified Explanations via Randomized Smoothing

Alaa Anani; Tobias Lorenz; Mario Fritz; Bernt Schiele

Pixel-level Certified Explanations via Randomized Smoothing Proceedings Article

In: International Conference on Machine Learning (ICML), 2025.

Exploring the Potential of LLMs for Code Deobfuscation

David Beste; Grégoire Menguy; Hossein Hajipour; Mario Fritz; Antonio Emanuele; Sébastien Bardin; Thorsten Holz; Thorsten Eisenhofer; Lea Schönherr

Exploring the Potential of LLMs for Code Deobfuscation Proceedings Article

In: SIG SIDAR Conference on Detection of Intrusions and Malware & Vulnerability Assessment (DIMVA), 2025.

ProtocolLLM: RTL Benchmark for SystemVerilog Generation of Communication Protocols

ProtocolLLM: RTL Benchmark for SystemVerilog Generation of Communication Protocols

Certifiably robust malware detectors by design

DocMIA: Document-Level Membership Inference Attacks against DocVQA Models

Khanh Nguyen; Raouf Kerkouche; Mario Fritz; Dimosthenis Karatzas

DocMIA: Document-Level Membership Inference Attacks against DocVQA Models Proceedings Article

In: Thirteenth International Conference on Learning Representations (ICLR), 2025.

Can LLMs Separate Instructions From Data? And What Do We Even Mean By That?

Get my drift? Catching LLM Task Drift with Activation Deltas

Medical Multimodal Model Stealing Attacks via Adversarial Domain Alignment

Proceedings

3D-Sec'25: The 1st ACM Workshop on Deepfake, Deception, and Disinformation Security

CSCS '25 - Cyber Security in CarS Workshop

Technical Reports

Toward Quantitative Modeling of Cybersecurity Risks Due to AI Misuse

Steve Barrett; Malcolm Murray; Otter Quarks; Matthew Smith; Jakub Kryś; Siméon Campos; Alejandro Tlaie Boria; Chloé Touzet; Sevan Hayrapet; Fred Heiding; Omer Nevo; Adam Swanda; Jair Aguirre; Asher Brass Gershovich; Eric Clay; Ryan Fetterman; Mario Fritz; Marc Juarez; Vasilios Mavroudis; Henry Papadatos

Toward Quantitative Modeling of Cybersecurity Risks Due to AI Misuse Technical Report

2025.

DP-SNP-TIHMM: Differentially Private, Time-Inhomogeneous Hidden Markov Models for Synthesizing Genome-Wide Association Datasets

In-Context Experience Replay Facilitates Safety Red-Teaming of Text-to-Image Diffusion Models

2024

Journal Articles

B-cos Alignment for Inherently Interpretable CNNs and Vision Transformers

A Unified View of Differentially Private Deep Generative Modeling

Proceedings Articles

Language Models as Zero-shot Lossless Gradient Compressors: Towards General Neural Parameter Prior Models

Cooperation, Competition, and Maliciousness: LLM-Stakeholders Interactive Negotiation

Dataset and Lessons Learned from the 2024 SaTML LLM Capture-the-Flag Competition

Edoardo Debenedetti; Javier Rando; Daniel Paleka; Silaghi Fineas Florin; Dragos Albastroiu; Niv Cohen; Yuval Lemberg; Reshmi Ghosh; Rui Wen; Ahmed Salem; Giovanni Cherubin; Santiago Zanella-Beguelin; Robin Schmid; Victor Klemm; Takahiro Miki; Chenhao Li; Stefan Kraft; Mario Fritz; Florian Tramèr; Sahar Abdelnabi; Lea Schönherr

Dataset and Lessons Learned from the 2024 SaTML LLM Capture-the-Flag Competition Proceedings Article

In: Neural Information Processing Systems (NeurIPS), 2024.

LLM2Swarm: Robot Swarms that Responsively Reason, Plan, and Collaborate through LLMs

Hypothesizing Missing Causal Variables with LLMs

CausalGraph2LLM: Evaluating LLMs for Causal Queries

LLM Task Interference: An Initial Study on the Impact of Task-Switch in Conversational History

Inside the Black Box: Detecting Data Leakage in Pre-trained Language Encoders

FullCert: Deterministic End-to-End Certification for Training and Inference of Neural Networks

Privacy-Aware Document Visual Question Answering

Rubèn Pérez Tito; Khanh Nguyen; Marlon Tobaben; Raouf Kerkouche; Mohamed Ali Souibgui; Gangsoo Jung; Joonas Jälkö; Vincent Poulain D'Andecy; Aurelie Joseph; Lei Kang; Ernest Valveny; Antti Honkela; Mario Fritz; Dimosthenis Karatzas

Privacy-Aware Document Visual Question Answering Proceedings Article

In: International Conference on Document Analysis and Recognition (ICDAR), 2024.

SecurityNet: Assessing Machine Learning Vulnerabilities on Public Models

Stealthy Imitation: Reward-guided Environment-free Policy Stealing

Adaptive Hierarchical Certification for Segmentation using Randomized Smoothing

MultiMax: Sparse and Multi-Modal Attention Learning

Towards Biologically Plausible and Private Gene Expression Data Generation

FedLAP-DP: Federated Learning by Sharing Differentially Private Loss Approximations

Tell me what you like and I know what you will share: Topical interest influences behavior toward news from high and low credible sources

SimSCOOD: Systematic Analysis of Out-of-Distribution Generalization in Fine-tuned Source Code Models

PoLLMgraph: Unraveling Hallucinations in Large Language Models via State Transition Dynamics

Can LLMs Separate Instructions From Data? And What Do We Even Mean By That?

LLM-Deliberation: Evaluating LLMs with Interactive Multi-Agent Negotiation Games

On Adversarial Training without Perturbing All Examples

CodeLMSec Benchmark: Systematically Evaluating and Finding Security Vulnerabilities in Black-Box Code Language Models

Miscellaneous

Fundamental Risks in the Current Deployment of General-Purpose AI Models: What Have We (Not) Learnt From Cybersecurity?

ELSA Strategic Research Agenda: Facing the Grand Challenges of Secure and Safe AI

Technical Reports

LLM4GRN: Discovering Causal Gene Regulatory Networks with LLMs – Evaluation through Synthetic Data Generation

HexaCoder: Secure Code Generation via Oracle-Guided Synthetic Training Data

2023

Journal Articles

Optimising for Interpretability: Convolutional Dynamic Alignment Networks

Proceedings Articles

Not what you've signed up for: Compromising Real-World LLM-Integrated Applications with Indirect Prompt Injection

Certifiers Make Neural Networks Vulnerable to Availability Attacks

Certified Robust Models with Slack Control and Large Lipschitz Constants

Fact-Saboteurs: A Taxonomy of Evidence Manipulation Attacks against Fact-Verification Systems

Compromising LLMs: The Advent of AI Malware

UnGANable: Defending Against GAN-based Face Manipulation

From Attachments to SEO: Click Here to Learn More about Clickbait PDFs!

Client-specific Property Inference against Secure Aggregation in Federated Learning

Proceedings

Proceedings of the 7th ACM Computer Science in Cars Symposium, CSCS 2023, Darmstadt, Germany, 5 December 2023

Technical Reports

2022

Journal Articles

Understanding Utility and Privacy of Demographic Data in Education Technology by Causal Analysis and Adversarial-Censoring

Proceedings Articles

Private Set Generation with Discriminative Information

ML-Doctor: Holistic Risk Assessment of Inference Attacks Against Machine Learning Models

ProgFed: Effective, Communication, and Computation Efficient Federated Learning by Progressive Training

B-cos Networks: Alignment is All We Need for Interpretability

Open-Domain, Content-based, Multi-modal Fact-checking of Out-of-Context Images via Online Resources

Responsible Disclosure of Generative Models Using Scalable Fingerprinting

RelaxLoss: Defending Membership Inference Attacks without Losing Utility

Dingfan Chen; Ning Yu; Mario Fritz

RelaxLoss: Defending Membership Inference Attacks without Losing Utility Proceedings Article

In: International Conference on Representation Learning (ICLR), 2022.

Practical Challenges in Differentially-Private Federated Survival Analysis of Medical Data

2021

Journal Articles

Semantic Bottlenecks: Quantifying and Improving Inspectability of Deep Representations

Privacy considerations for sharing genomics data

Proceedings Articles

Artificial Fingerprinting for Generative Models: Rooting Deepfake Attribution in Training Data

Dual Contrastive Loss and Attention for GANs

Beyond the Spectrum: Detecting Deepfakes via Re-Synthesis

Convolutional Dynamic Alignment Networks for Interpretable Classifications

Euro-PVI: Pedestrian Vehicle Interactions in Dense Urban Centers

Hijack-GAN: Unintended-Use of Pretrained, Black-Box GANs

Adversarial Watermarking Transformer: Towards Tracing Text Provenance with Data Hiding

Future Moment Assessment for Action Query

Technical Reports

Backdoor Attacks on Network Certification via Data Poisoning

ML-Doctor: Holistic Risk Assessment of Inference Attacks Against Machine Learning Models

Workshops

"What's in the box?!": Deflecting Adversarial Attacks by Randomly Deploying Adversarially-Disjoint Models

SampleFix: Learning to Generate Functionally Diverse Fixes

IReEn: Iterative Reverse-Engineering of Black-Box Functions via Neural Program Synthesis

InfoScrub: Towards Attribute Privacy by Targeted Obfuscation

MLCapsule: Guarded Offline Deployment of Machine Learning as a Service

2020

Journal Articles

Person Recognition in Personal Photo Collections

Deep Gaze Pooling: Inferring and Visually Decoding Search Intents From Human Gaze Fixations

Proceedings Articles

GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generators

GAN-Leaks: A Taxonomy of Membership Inference Attacks against GANs

VisualPhishNet: Zero-Day Phishing Website Detection by Visual Similarity

Haar Wavelet based Block Autoregressive Flows for Trajectories

Long-Tailed Recognition Using Class-Balanced Experts

Semantic Bottlenecks: Quantifying & Improving Inspectability of Deep Representations

Towards Automated Testing and Robustification by Semantic Adversarial Data Generation

Inclusive GAN: Improving Data and Minority Coverage in Generative Models

Segmentations-Leak: Membership Inference Attacks and Defenses in Semantic Image Segmentation

Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learning

Towards Causal VQA: Revealing and Reducing Spurious Correlations by Invariant and Covariant Semantic Editing

Normalizing Flows with Multi-scale Autoregressive Priors

Automatically Detecting Bystanders in Photos to Reduce Privacy Risks

Prediction Poisoning: Utility-Constrained Defenses Against Model Stealing Attacks

Technical Reports

CosSGD: Nonlinear Quantization for Communication-efficient Federated Learning

Responsible Disclosure of Generative Models Using Scalable Fingerprinting

Hijack-GAN: Unintended-Use of Pretrained, Black-Box GANs

Adversarial Watermarking Transformer: Towards Tracing Text Provenance with Data Hiding

Black-Box Watermarking for Generative Adversarial Networks

IReEn: Iterative Reverse-Engineering of Black-Box Functions via Neural Program Synthesis

GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generators

InfoScrub: Towards Attribute Privacy by Targeted Obfuscation

Inclusive GAN: Improving Data and Minority Coverage in Generative Models

Long-Tailed Recognition Using Class-Balanced Experts

Workshops

SampleFix: Learning to Correct Programs by Sampling Diverse Fixes

IReEn: Iterative Reverse-Engineering of Black-Box Functions via Neural Program Synthesis

Haar Wavelet based Block Autoregressive Flows for Trajectories

Body Shape Privacy in Images: Understanding Privacy and Preventing Automatic Shape Extraction

Synthetic Convolutional Features for Improved Semantic Segmentation

2019

Journal Articles

MPIIGaze: Real-World Dataset and Deep Appearance-Based Gaze Estimation

Book Sections

Towards reverse-engineering black-box neural networks

Proceedings Articles

Attributing Fake Images to GANs: Learning and Analyzing GAN Fingerprints

Deep Appearance Maps

Maxim Maximov; Tobias Ritschel; Laura Leal-Taixe; Mario Fritz

Deep Appearance Maps Proceedings Article

In: International Conference on Computer Vision (ICCV), 2019.

Knockoff Nets: Stealing Functionality of Black-Box Models

Not Using the Car to See the Sidewalk: Quantifying and Controlling the Effects of Context in Classification and Segmentation

Time-Conditioned Action Anticipation in One Shot

Qiuhong Ke; Mario Fritz; Bernt Schiele

Time-Conditioned Action Anticipation in One Shot Proceedings Article

In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2019.

Bayesian Prediction of Future Street Scenes using Synthetic Likelihoods

ML-Leaks: Model and Data Independent Membership Inference Attacks and Defenses on Machine Learning Models

Fashion is Taking Shape: Understanding Clothing Preference Based on Body Shape From Online Sources

Technical Reports

Segmentations-Leak: Membership Inference Attacks and Defenses in Semantic Image Segmentation

Towards Causal VQA: Revealing and Reducing Spurious Correlations by Invariant and Covariant Semantic Editing

"Best-of-Many-Samples" Distribution Matching

GAN-Leaks: A Taxonomy of Membership Inference Attacks against GANs

WhiteNet: Phishing Website Detection by Visual Whitelists

Conditional Flow Variational Autoencoders for Structured Sequence Prediction

Interpretability Beyond Classification Output: Semantic Bottleneck Networks

Prediction Poisoning: Utility-Constrained Defenses Against Model Stealing Attacks

SampleFix: Learning to Correct Programs by Sampling Diverse Fixes

Shape Evasion: Preventing Body Shape Inference of Multi-Stage Approaches

Learning Manipulation under Physics Constraints with Visual Perception

Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learning

Workshops

Differential Privacy Defenses and Sampling Attacks for Membership Inference

Updates-Leak: Data Set Inference and Reconstruction Attacks in Online Learning

Understanding and Recognizing Bystanders in Images for Privacy Protection

2018

Journal Articles

Advanced Steel Microstructural Classification by Deep Learning Methods

Reflectance and Natural Illumination from Single-Material Specular Objects Using Deep Learning

Stamatios Georgoulis; Konstantinos Rematas; Tobias Ritschel; Efstratios Gavves; Mario Fritz; Luc Van Gool; Tinne Tuytelaars

Reflectance and Natural Illumination from Single-Material Specular Objects Using Deep Learning Journal Article

In: Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2018.

Proceedings Articles

Adversarial Scene Editing: Automatic Object Removal from Weak Supervision

Diverse Conditional Image Generation by Stochastic Regression with Latent Drop-Out Codes

A Hybrid Model for Identity Obfuscation by Face Replacement

Answering Visual What-If Questions: From Actions to Predicted Scene Descriptions

Sequential Attacks on Agents for Long-Term Adversarial Goals

Edgar Tretschk; Seong Joon Oh; Mario Fritz

Sequential Attacks on Agents for Long-Term Adversarial Goals Proceedings Article

In: 2. ACM Computer Science in Cars Symposium -- Future Challenges in Artificial Intelligence & Security for Autonomous Vehicles, 2018.

A4NT: Author Attribute Anonymity by Adversarial Training of Neural Machine Translation

Accurate and Diverse Sampling of Sequences based on a “Best of Many” Sample Objective

Connecting Pixels to Privacy and Utility: Automatic Redaction of Private Information in Images

Natural and Effective Obfuscation by Head Inpainting

Disentangled Person Image Generation

Liqian Ma; Qianru Sun; Stamatios Georgoulis; Luc Van Gool; Bernt Schiele; Mario Fritz

Disentangled Person Image Generation Proceedings Article

In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018.

Long-Term On-Board Prediction of People in Traffic Scenes under Uncertainty

Towards Reverse-Engineering Black-Box Neural Networks

Long-Term Image Boundary Prediction

Apratim Bhattacharyya; Mateusz Malinowski; Bernt Schiele; Mario Fritz

Long-Term Image Boundary Prediction Proceedings Article

In: Association for the Advancement of Artificial Intelligence (AAAI), 2018.

Technical Reports

Not Using the Car to See the Sidewalk: Quantifying and Controlling the Effects of Context in Classification and Segmentation

Knockoff Nets: Stealing Functionality of Black-Box Models

Attributing Fake Images to GANs: Analyzing Fingerprints in Generated Images

MLCapsule: Guarded Offline Deployment of Machine Learning as a Service

Fashion is Taking Shape: Understanding Clothing Preference Based on Body Shape From Online Sources

ML-Leaks: Model and Data Independent Membership Inference Attacks and Defenses on Machine Learning Models

Adversarial Scene Editing: Automatic Object Removal from Weak Supervision

Sequential Attacks on Agents for Long-Term Adversarial Goals

Understanding and Controlling User Linkability in Decentralized Learning

A Hybrid Model for Identity Obfuscation by Face Replacement

Deep Appearance Maps

Maxim Maximov; Tobias Ritschel; Mario Fritz

Deep Appearance Maps Technical Report

arXiv:1804.00863 [cs.CV], 2018.

2017

Journal Articles

Ask Your Neurons: A Deep Learning Approach to Visual Question Answering

Novel Views of Objects from a Single Image

Konstantinos Rematas; Chuong Nguyen; Tobias Ritschel; Mario Fritz; Tinne Tuytelaars

Novel Views of Objects from a Single Image Journal Article

In: Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2017.

Proceedings Articles

Predicting the Category and Attributes of Visual Search Targets Using Deep Gaze Pooling

Towards a Visual Privacy Advisor: Understanding and Predicting Privacy Risks in Images

Speaking the Same Language: Matching Machine to Human Captions by Adversarial Training

Adversarial Image Perturbation for Privacy Protection -- A Game Theory Perspective

What Is Around The Camera?

Stamatios Georgoulis; Konstantinos Rematas; Tobias Ritschel; Mario Fritz; Tinne Tuytelaars; Luc Van Gool

What Is Around The Camera? Proceedings Article

In: IEEE International Conference on Computer Vision (ICCV), 2017.

Learning Dilation Factors for Semantic Segmentation of Street Scenes

STD2P: RGBD Semantic Segmentation Using Spatio-Temporal Data-Driven Pooling

A Domain Based Approach to Social Relation Recognition

Exploiting saliency for object segmentation from image level labels

It's Written All Over Your Face: Full-Face Appearance-Based Gaze Estimation

Visual Stability Prediction for Robotic Manipulation

Miscellaneous

Long-Term On-Board Prediction of Pedestrians in Traffic Scenes

From Understanding to Controlling Privacy against Automatic Person Identification in Social Media

Visual Stability Prediction and Its Application to Manipulation

Technical Reports

Disentangled Person Image Generation

Connecting Pixels to Privacy and Utility: Automatic Redaction of Private Information in Images

Natural and Effective Obfuscation by Head Inpainting

Long-Term On-Board Prediction of People in Traffic Scenes under Uncertainty

A4NT : Author Attribute Anonymity by Adversarial Training of Neural Machine Translation

Whitening Black-Box Neural Networks

Acquiring Target Stacking Skills by Goal-Parameterized Deep Reinforcement Learning

Person Recognition in Social Media Photos

Advanced Steel Microstructure Classification by Deep Learning Methods

Visual Decoding of Targets During Visual Search From Human Eye Fixations

Towards a Visual Privacy Advisor: Understanding and Predicting Privacy Risks in Images

Speaking the Same Language: Matching Machine to Human Captions by Adversarial Training

Adversarial Image Perturbation for Privacy Protection -- A Game Theory Perspective

Exploiting saliency for object segmentation from image level labels

2016

Proceedings Articles

Faceless Person Recognition; Privacy Implications in Social Media

VConv-DAE: Deep Volumetric Shape Learning Without Object Labels

Towards Segmenting Consumer Stereo Videos: Benchmark, Baselines and Ensembles

Mean Box Pooling: A Rich Image Representation and Output Embedding for the Visual Madlibs Task

Deep Reflectance Maps

Konstantinos Rematas; Tobias Ritschel; Mario Fritz; Efstratios Gavves; Tinne Tuytelaars

Deep Reflectance Maps Proceedings Article

In: IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016.

Multi-Cue Zero-Shot Learning with Strong Supervision

I-Pic: A Platform for Privacy-Compliant Image Capture

Paarijaat Aditya; Rijurekha Sen; Seong Joon Oh; Rodrigo Benenson; Bobby Bhattacharjee; Peter Druschel; Tongtong Wu; Mario Fritz; Bernt Schiele

I-Pic: A Platform for Privacy-Compliant Image Capture Proceedings Article

In: The 14th International Conference on Mobile Systems, Applications, and Services (MobiSys'16), Singapore, 2016.

Learning to Select Long Track Features for Structure-From-Motion & Visual SLAM

Jonas Scheer; Mario Fritz; Oliver Grau

Learning to Select Long Track Features for Structure-From-Motion & Visual SLAM Proceedings Article

In: German Conference on Pattern Recognition (GCPR), 2016.

Contextual Media Retrieval Using Natural Language Queries

Recognition of Ongoing Complex Activities by Sequence Prediction over a Hierarchical Label Space

Miscellaneous

Visual Stability Prediction and Its Application to Manipulation

Long Term Boundary Extrapolation for Deterministic Motion

Faceless Person Recognition; Privacy Implications in Social Media

Ask Your Neurons Again: Analysis of Deep Methods with Global Image Representation

PhD Theses

Bayesian Non-Parametrics for Multi-Modal Segmentation

Technical Reports

Predicting the Category and Attributes of Visual Search Targets Using Deep Gaze Pooling

Long-Term Image Boundary Extrapolation

Natural Illumination from Multiple Materials Using Deep Learning

It's Written All Over Your Face: Full-Face Appearance-Based Gaze Estimation

Tutorial on Answering Questions about Images with Deep Learning

Visual Stability Prediction and Its Application to Manipulation

Spatio-Temporal Image Boundary Extrapolation

Ask Your Neurons: A Deep Learning Approach to Visual Question Answering

VConv-DAE: Deep Volumetric Shape Learning Without Object Labels

RGBD Semantic Segmentation Using Spatio-Temporal Data-Driven Pooling

To Fall Or Not To Fall: A Visual Approach to Physical Stability Prediction

DeLight-Net: Decomposing Reflectance Maps into Specular Materials and Natural Illumination

Multi-Cue Zero-Shot Learning with Strong Supervision

Novel Views of Objects from a Single Image

Contextual Media Retrieval Using Natural Language Queries

2015

Journal Articles

Learning to detect visual grasp affordance

Proceedings Articles

Ask Your Neurons: A Neural-based Approach to Answering Questions about Images

See the Difference: Direct Pre-Image Reconstruction and Pose Estimation by Differentiating HOG

Person Recognition in Personal Photo Collections

Teaching Robots the Use of Human Tools from Demonstration with Non-Dexterous End-Effectors

Appearance-based gaze estimation in the wild

Prediction of search targets from fixations in open-world settings

Joint Segmentation and Activity Discovery using Semantic and Temporal Priors

Hard to Cheat: A Turing Test based on Answering Questions about Images

Masters Theses

Contextual Media Retrieval Using Natural Language Queries

Sreyasi Nag Chowdhury

Contextual Media Retrieval Using Natural Language Queries Masters Thesis

Saarland University, 2015.

Miscellaneous

Bridging the Gap Between Synthetic and Real Data

Technical Reports

Deep Reflectance Maps

Konstantinos Rematas; Tobias Ritschel; Mario Fritz; Efstratios Gavves; Tinne Tuytelaars

Deep Reflectance Maps Technical Report

arXiv:1511.04384 [cs.CV], 2015.

Person Recognition in Personal Photo Collections

Appearance-based gaze estimation in the wild

Prediction of search targets from fixations in open-world settings

Ask Your Neurons: A Neural-based Approach to Answering Questions about Images

See the Difference: Direct Pre-Image Reconstruction and Pose Estimation by Differentiating HOG

GazeDPM: Early Integration of Gaze Information in Deformable Part Models

2014

Proceedings Articles

A Multi-World Approach to Question Answering about Real-World Scenes based on Uncertain Input

Anytime Recognition of Objects and Scenes

Image-based Synthesis and Re-Synthesis of Viewpoints Guided by 3D Models

Towards a Visual Turing Challenge

Object Disambiguation for Augmented Reality Applications

Ubic: Bridging the Gap Between Digital Cryptography and the Physical World

Scene Segmentation in Adverse Vision Conditions

Learning Multi-Scale Representations for Material Classification

Technical Reports

A Pooling Approach to Modelling Spatial Relations for Image Retrieval and Annotation

Learning Multi-Scale Representations for Material Classification

Ubic: Bridging the gap between digital cryptography and the physical world

2013

Book Sections

A Category-Level 3D Object Dataset: Putting the Kinect to Work

Allison Janoch; Sergey Karayev; Yangqing Jia; Jonathan T. Barron; Mario Fritz; Kate Saenko; Trevor Darrell

A Category-Level 3D Object Dataset: Putting the Kinect to Work Book Section

In: Fossati, Andrea; Gall, Juergen; Grabner, Helmut; Ren, Xiaofeng; Konolige, Kurt (Ed.): Consumer Depth Cameras for Computer Vision, Springer London, 2013.

Proceedings Articles

Sequential Bayesian Model Update under Structured Scene Prior for Semantic Road Scenes Labeling

Learning Smooth Pooling Regions for Visual Recognition

Dynamic Feature Selection for Classification on a Budget

Multi-Class Video Co-Segmentation with a Generative Multi-Video Model

Masters Theses

Scene Segmentation in Adverse Vision Conditions

Multi-Scale Feature Learning for Material Recognition

Wenbin Li

Multi-Scale Feature Learning for Material Recognition Masters Thesis

Saarland University, 2013.

Technical Reports

Learnable Pooling Regions for Image Classification

2012

Journal Articles

A Geometric Approach To Robotic Laundry Folding

Proceedings Articles

Kernel Density Topic Models: Visual Topics Without Visual Words

Active Metric Learning for Object Recognition

Semi-Supervised Learning on a Budget: Scaling up to Large Datasets

The Pooled NBNN Kernel: Beyond Image-to-Class and Image-to-Image

Timely Object Recognition

Sergey Karayev; Tobias Baumgartner; Mario Fritz; Trevor Darrell

Timely Object Recognition Proceedings Article

In: Advances in Neural Information Processing Systems (NIPS), 2012.

Sparselet Models for Efficient Multiclass Object Detection

Recognizing Materials from Virtual Examples

RALF: A Reinforced Active Learning Formulation for Object Class Recognition

2011

Proceedings Articles

Parameterized Shape Models for Clothing

A Probabilistic Model for Recursive Factorized Image Features

Pick your Neighborhood -- Improving Labels and Neighborhood Structure for Label Propagation

Improving the Kinect by Cross-Modal Stereo

Practical 3-D Object Detection Using Category and Instance-level Appearance Models

Perception for the Manipulation of Socks

Ping Chuan Wang; Stephen Miller; Mario Fritz; Trevor Darrell; Pieter Abbeel

Perception for the Manipulation of Socks Proceedings Article

In: IEEE International Conference on Intelligent Robots and Systems (IROS), 2011.

The NBNN kernel

Tinne Tuytelaars; Mario Fritz; Kate Saenko; Trevor Darrell

The NBNN kernel Proceedings Article

In: IEEE International Conference on Computer Vision (ICCV), 2011.

I spy with my little eye: Learning Optimal Filters for Cross-Modal Stereo under Projected Patterns

Visual Grasp Affordances From Appearance-Based Cues

A Category-Level 3-D Object Dataset: Putting the Kinect to Work

Masters Theses

Optimization Algorithms in the Reconstruction of MR Images: A Comparative Study

2010

Journal Articles

Tutor-based learning of visual categories using different levels of supervision

Classifying materials in the real world

Book Sections

Categorical Perception

Mario Fritz; Mykhaylo Andriluka; Sanja Fidler; Michael Stark; Ales Leonardis; Bernt Schiele

Categorical Perception Book Section

In: Cognitive Systems, Springer, 2010.

Multi-Modal Learning

Danijel Skocaj; Matej Kristan; Alen Vrecko; Ales Leonardis; Mario Fritz; Michael Stark; Bernt Schiele; Somboon Hongeng; Jeremy L. Wyatt

Multi-Modal Learning Book Section

In: Cognitive Systems, Springer, 2010.

Size Matters: Metric Visual Search Constraints from Monocular Metadata

Proceedings Articles

Adapting visual category models to new domains

2009

Book Sections

Towards Integration of Different Paradigms in Modeling, Representation and Learning of Visual Categories

Proceedings Articles

Discriminative Structure Learning of Hierarchical Representations for Object Detection

An Additive Latent Feature Model for Transparent Object Recognition

PhD Theses

Modeling, Representing and Learning of Visual Categories

Mario Fritz

Modeling, Representing and Learning of Visual Categories PhD Thesis

TU Darmstadt, 2009.

Proceedings

Computer Vision Systems, 7th International Conference on Computer Vision Systems, ICVS 2009, Liege, Belgium, October 13-15, 2009, Proceedings

Mario Fritz; Bernt Schiele; Justus H. Piater (Ed.)

Computer Vision Systems, 7th International Conference on Computer Vision Systems, ICVS 2009, Liege, Belgium, October 13-15, 2009, Proceedings Proceedings

Springer, vol. 5815, 2009, ISBN: 978-3-642-04666-7.

2008

Proceedings Articles

Discovery of activity patterns using topic models

Hierarchical Support Vector Random Fields: Joint Training to Combine Local and Global Features

Decomposition, Discovery and Detection of Visual Categories Using Topic Models

2007

Proceedings Articles

Towards Robust Pedestrian Detection in Crowded Image Sequences

Cross-Modal Learning of Visual Categories using Different Levels of Supervision

2006

Proceedings Articles

The 2005 PASCAL Visual Object Classes Challenge

Mark Everingham; Andrew Zisserman; Christopher K. I. Williams; Luc Van Gool; Moray Allan; Christopher M. Bishop; Olivier Chapelle; Navneet Dalal; Thomas Deselaers; Gyuri Dork'o; Stefan Duffner; Jan Eichhorn; Jason D. R. Farquhar; Mario Fritz; Christophe Garcia; Tom Griffiths; Fr'ed'eric Jurie; Thomas Keysers; Markus Koskela; Jorma Laaksonen; Larlus, Diane; Bastian Leibe; Hongying Meng; Hermann Ney; Bernt Schiele; Corde

The 2005 PASCAL Visual Object Classes Challenge Proceedings Article

In: Selected Proceedings of the first PASCAL Challenges Workshop, 2006.

Towards Unsupervised Discovery of Visual Categories

2005

Proceedings Articles

Integrating Representative and Discriminant Models for Object Category Detection

2004

Proceedings Articles

On the Significance of Real-World Conditions for Material Classification

Miscellaneous

Categorization by Local Information Using Support Vector Machines

2002

Proceedings Articles

Object Tracking and Pose Estimation Using Light-Field Object Models

Miscellaneous

3D Objektverfolgung mit Lichtfeldern (3D object tracking using light-fields)

Mario Fritz

3D Objektverfolgung mit Lichtfeldern (3D object tracking using light-fields) Miscellaneous

2002, (student thesis).

Read the original on cispa.saarland ↗