no code implementations • ECCV 2020 • Rakshith Shetty, Mario Fritz, Bernt Schiele
Constrained adversarial optimization of object appearance through this synthesizer produces rare/difficult versions of an object which fool the target object detector.
no code implementations • 9 Jun 2025 • Arnav Sheth, Ivaxi Sheth, Mario Fritz
Recent advances in Large Language Models (LLMs) have shown promising capabilities in generating code for general-purpose programming languages.
no code implementations • 28 Feb 2025 • Ruta Binkyte, Ivaxi Sheth, Zhijing Jin, Mohammad Havaei, Bernhard Schölkopf, Mario Fritz
Ensuring trustworthiness in machine learning (ML) systems is crucial as they become increasingly embedded in high-stakes domains.
no code implementations • 27 Feb 2025 • Jan Wehner, Sahar Abdelnabi, Daniel Tan, David Krueger, Mario Fritz
We present the first comprehensive survey of RepE for LLMs, reviewing the rapidly growing literature to address key questions: What RepE methods exist and how do they differ?
1 code implementation • 18 Feb 2025 • Yuxuan Zhou, Heng Li, Zhi-Qi Cheng, Xudong Yan, Mario Fritz, Margret Keuper
Label Smoothing (LS) is widely adopted to curb overconfidence in neural network predictions and enhance generalization.
no code implementations • 6 Feb 2025 • Ivaxi Sheth, Jan Wehner, Sahar Abdelnabi, Ruta Binkyte, Mario Fritz
AI advancements have been significantly driven by a combination of foundation models and curiosity-driven learning aimed at increasing capability and adaptability.
1 code implementation • 6 Feb 2025 • Khanh Nguyen, Raouf Kerkouche, Mario Fritz, Dimosthenis Karatzas
Document Visual Question Answering (DocVQA) has introduced a new paradigm for end-to-end document understanding, and quickly became one of the standard benchmarks for multimodal LLMs.
no code implementations • 4 Feb 2025 • Yaling Shen, Zhixiong Zhuang, Kun Yuan, Maria-Irina Nicolae, Nassir Navab, Nicolas Padoy, Mario Fritz
Experiments on the IU X-RAY and MIMIC-CXR radiology datasets demonstrate that Adversarial Domain Alignment enables attackers to steal the medical MLLM without any access to medical data.
no code implementations • 10 Jan 2025 • Sarath Sivaprasad, Dmitry Kangin, Plamen Angelov, Mario Fritz
Aligning machine representations with human understanding is key to improving interpretability of machine learning (ML) models.
no code implementations • 19 Dec 2024 • Mario Fritz
General Purpose AI - such as Large Language Models (LLMs) - have seen rapid deployment in a wide range of use cases.
no code implementations • 13 Dec 2024 • Tobias Lorenz, Marta Kwiatkowska, Mario Fritz
A key element to make this computation feasible is to relax the reachable parameter set to a convex set between training iterations.
1 code implementation • 3 Dec 2024 • Tejumade Afonja, Hui-Po Wang, Raouf Kerkouche, Mario Fritz
To overcome this, we propose DP-2Stage, a two-stage fine-tuning framework for differentially private tabular data generation.
no code implementations • 25 Nov 2024 • Zhi-Yi Chin, Mario Fritz, Pin-Yu Chen, Wei-Chen Chiu
Text-to-image (T2I) models have shown remarkable progress, but their potential to generate harmful content remains a critical concern in the ML community.
no code implementations • 6 Nov 2024 • 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 Llados, Ernest Valveny, Antti Honkela, Mario Fritz, Dimosthenis Karatzas
The Privacy Preserving Federated Learning Document VQA (PFL-DocVQA) competition challenged the community to develop provably private and communication-efficient solutions in a federated setting for a real-life use case: invoice processing.
1 code implementation • 21 Oct 2024 • Ivaxi Sheth, Bahare Fatemi, Mario Fritz
In this paper, we propose a comprehensive benchmark, \emph{CausalGraph2LLM}, encompassing a variety of causal graph settings to assess the causal graph understanding capability of LLMs.
no code implementations • 21 Oct 2024 • Tejumade Afonja, Ivaxi Sheth, Ruta Binkyte, Waqar Hanif, Thomas Ulas, Matthias Becker, Mario Fritz
Gene regulatory networks (GRNs) represent the causal relationships between transcription factors (TFs) and target genes in single-cell RNA sequencing (scRNA-seq) data.
1 code implementation • 26 Sep 2024 • Hui-Po Wang, Mario Fritz
In this work, we demonstrate the potential of large language models (LLMs) to act as gradient priors in a zero-shot setting.
1 code implementation • 10 Sep 2024 • Hossein Hajipour, Lea Schönherr, Thorsten Holz, Mario Fritz
The data synthesis pipeline generates pairs of vulnerable and fixed codes for specific Common Weakness Enumeration (CWE) types by utilizing a state-of-the-art LLM for repairing vulnerable code.
1 code implementation • 4 Sep 2024 • Ivaxi Sheth, Sahar Abdelnabi, Mario Fritz
Motivated by the scientific discovery process, in this work, we formulate a novel task where the input is a partial causal graph with missing variables, and the output is a hypothesis about the missing variables to complete the partial graph.
1 code implementation • 24 Aug 2024 • Yuxuan Zhou, Margret Keuper, Mario Fritz
Sampling-based decoding strategies have been widely adopted for Large Language Models (LLMs) in numerous applications, targeting a balance between diversity and quality via temperature tuning and tail truncation.
no code implementations • 20 Aug 2024 • Yuan Xin, Zheng Li, Ning Yu, Dingfan Chen, Mario Fritz, Michael Backes, Yang Zhang
Despite being prevalent in the general field of Natural Language Processing (NLP), pre-trained language models inherently carry privacy and copyright concerns due to their nature of training on large-scale web-scraped data.
1 code implementation • 17 Jun 2024 • Tobias Lorenz, Marta Kwiatkowska, Mario Fritz
In this work, we present FullCert, the first end-to-end certifier with sound, deterministic bounds, which proves robustness against both training-time and inference-time attacks.
1 code implementation • 12 Jun 2024 • 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
To study this problem, we organized a capture-the-flag competition at IEEE SaTML 2024, where the flag is a secret string in the LLM system prompt.
1 code implementation • 3 Jun 2024 • Yuxuan Zhou, Mario Fritz, Margret Keuper
Yet, as a smooth approximation to the Argmax function, a significant amount of probability mass is distributed to other, residual entries, leading to poor interpretability and noise.
1 code implementation • 2 Jun 2024 • Sahar Abdelnabi, Aideen Fay, Giovanni Cherubin, Ahmed Salem, Mario Fritz, Andrew Paverd
We study LLM activations as a solution to detect task drift, showing that activation deltas - the difference in activations before and after processing external data - are strongly correlated with this phenomenon.
no code implementations • 11 May 2024 • Zhixiong Zhuang, Maria-Irina Nicolae, Mario Fritz
Deep reinforcement learning policies, which are integral to modern control systems, represent valuable intellectual property.
1 code implementation • 6 Apr 2024 • Derui Zhu, Dingfan Chen, Qing Li, Zongxiong Chen, Lei Ma, Jens Grossklags, Mario Fritz
Despite tremendous advancements in large language models (LLMs) over recent years, a notably urgent challenge for their practical deployment is the phenomenon of hallucination, where the model fabricates facts and produces non-factual statements.
1 code implementation • 11 Mar 2024 • Egor Zverev, Sahar Abdelnabi, Soroush Tabesh, Mario Fritz, Christoph H. Lampert
We also present a new dataset, SEP, that allows estimating the measure for real-world models.
1 code implementation • 28 Feb 2024 • Akash Gupta, Ivaxi Sheth, Vyas Raina, Mark Gales, Mario Fritz
With the recent emergence of powerful instruction-tuned large language models (LLMs), various helpful conversational Artificial Intelligence (AI) systems have been deployed across many applications.
no code implementations • 16 Feb 2024 • Sarath Sivaprasad, Pramod Kaushik, Sahar Abdelnabi, Mario Fritz
We study this sampling behavior and show that this underlying heuristics resembles that of human decision-making: comprising a descriptive component (reflecting statistical norm) and a prescriptive component (implicit ideal encoded in the LLM) of a concept.
1 code implementation • 13 Feb 2024 • Alaa Anani, Tobias Lorenz, Bernt Schiele, Mario Fritz
Certification for machine learning is proving that no adversarial sample can evade a model within a range under certain conditions, a necessity for safety-critical domains.
1 code implementation • 7 Feb 2024 • Dingfan Chen, Marie Oestreich, Tejumade Afonja, Raouf Kerkouche, Matthias Becker, Mario Fritz
In this paper, we initiate a systematic analysis of how DP generative models perform in their natural application scenarios, specifically focusing on real-world gene expression data.
1 code implementation • 15 Dec 2023 • Rubèn Tito, Khanh Nguyen, Marlon Tobaben, Raouf Kerkouche, Mohamed Ali Souibgui, Kangsoo Jung, Joonas Jälkö, Vincent Poulain D'Andecy, Aurelie Joseph, Lei Kang, Ernest Valveny, Antti Honkela, Mario Fritz, Dimosthenis Karatzas
We employ a federated learning scheme, that reflects the real-life distribution of documents in different businesses, and we explore the use case where the data of the invoice provider is the sensitive information to be protected.
no code implementations • 1 Oct 2023 • Sarath Sivaprasad, Mario Fritz
We propose a novel approach to AD using explainability to capture such novel features as unexplained observations in the input space.
2 code implementations • 29 Sep 2023 • Sahar Abdelnabi, Amr Gomaa, Sarath Sivaprasad, Lea Schönherr, Mario Fritz
The fundamental task of negotiation spans many key features of communication, such as cooperation, competition, and manipulation potentials.
no code implementations • 27 Sep 2023 • Dingfan Chen, Raouf Kerkouche, Mario Fritz
The availability of rich and vast data sources has greatly advanced machine learning applications in various domains.
1 code implementation • 12 Sep 2023 • Max Losch, David Stutz, Bernt Schiele, Mario Fritz
In this paper, we propose a Calibrated Lipschitz-Margin Loss (CLL) that addresses this issue and improves certified robustness by tackling two problems: Firstly, commonly used margin losses do not adjust the penalties to the shrinking output distribution; caused by minimizing the Lipschitz constant $K$.
no code implementations • 16 Jul 2023 • Tejumade Afonja, Dingfan Chen, Mario Fritz
The potential of realistic and useful synthetic data is significant.
1 code implementation • 19 Jun 2023 • Moritz Böhle, Navdeeppal Singh, Mario Fritz, Bernt Schiele
We present a new direction for increasing the interpretability of deep neural networks (DNNs) by promoting weight-input alignment during training.
1 code implementation • 7 Mar 2023 • Raouf Kerkouche, Gergely Ács, Mario Fritz
We formulate an optimization problem across different rounds in order to infer a tested property of every client from the output of the linear models, for example, whether they have a specific sample in their training data (membership inference) or whether they misbehave and attempt to degrade the performance of the common model by poisoning attacks.
2 code implementations • 23 Feb 2023 • Kai Greshake, Sahar Abdelnabi, Shailesh Mishra, Christoph Endres, Thorsten Holz, Mario Fritz
Large Language Models (LLMs) are increasingly being integrated into various applications.
no code implementations • 15 Feb 2023 • Derui Zhu, Dingfan Chen, Jens Grossklags, Mario Fritz
In recent years, diffusion models have achieved tremendous success in the field of image generation, becoming the stateof-the-art technology for AI-based image processing applications.
no code implementations • 8 Feb 2023 • Hossein Hajipour, Keno Hassler, Thorsten Holz, Lea Schönherr, Mario Fritz
We evaluate the effectiveness of our approach by examining code language models in generating high-risk security weaknesses.
3 code implementations • 2 Feb 2023 • Hui-Po Wang, Dingfan Chen, Raouf Kerkouche, Mario Fritz
Our formulation involves clients synthesizing a small set of samples that approximate local loss landscapes by simulating the gradients of real images within a local region.
no code implementations • 20 Jan 2023 • Moritz Böhle, Mario Fritz, Bernt Schiele
Transformers increasingly dominate the machine learning landscape across many tasks and domains, which increases the importance for understanding their outputs.
2 code implementations • 7 Nov 2022 • Dingfan Chen, Raouf Kerkouche, Mario Fritz
Differentially private data generation techniques have become a promising solution to the data privacy challenge -- it enables sharing of data while complying with rigorous privacy guarantees, which is essential for scientific progress in sensitive domains.
1 code implementation • 10 Oct 2022 • Hossein Hajipour, Ning Yu, Cristian-Alexandru Staicu, Mario Fritz
In this paper, we contribute the first systematic approach that simulates various OOD scenarios along different dimensions of source code data properties and study the fine-tuned model behaviors in such scenarios.
1 code implementation • 3 Oct 2022 • Zheng Li, Ning Yu, Ahmed Salem, Michael Backes, Mario Fritz, Yang Zhang
Extensive experiments on four popular GAN models trained on two benchmark face datasets show that UnGANable achieves remarkable effectiveness and utility performance, and outperforms multiple baseline methods.
1 code implementation • 7 Sep 2022 • Sahar Abdelnabi, Mario Fritz
In this work, we assume an adversary that automatically tampers with the online evidence in order to disrupt the fact-checking model via camouflaging the relevant evidence or planting a misleading one.
1 code implementation • ICLR 2022 • Dingfan Chen, Ning Yu, Mario Fritz
As a long-term threat to the privacy of training data, membership inference attacks (MIAs) emerge ubiquitously in machine learning models.
1 code implementation • CVPR 2022 • Moritz Böhle, Mario Fritz, Bernt Schiele
We present a new direction for increasing the interpretability of deep neural networks (DNNs) by promoting weight-input alignment during training.
no code implementations • 8 Feb 2022 • Shadi Rahimian, Raouf Kerkouche, Ina Kurth, Mario Fritz
Survival analysis or time-to-event analysis aims to model and predict the time it takes for an event of interest to happen in a population or an individual.
no code implementations • CVPR 2022 • Sahar Abdelnabi, Rakibul Hasan, Mario Fritz
Misinformation is now a major problem due to its potential high risks to our core democratic and societal values and orders.
2 code implementations • 11 Oct 2021 • Hui-Po Wang, Sebastian U. Stich, Yang He, Mario Fritz
Federated learning is a powerful distributed learning scheme that allows numerous edge devices to collaboratively train a model without sharing their data.
1 code implementation • 27 Sep 2021 • Moritz Böhle, Mario Fritz, Bernt Schiele
As a result, CoDA Nets model the classification prediction through a series of input-dependent linear transformations, allowing for linear decomposition of the output into individual input contributions.
no code implementations • 25 Aug 2021 • Tobias Lorenz, Marta Kwiatkowska, Mario Fritz
While this is a key concept towards safe and secure AI, we show for the first time that this approach comes with its own security risks, as such fallback strategies can be deliberately triggered by an adversary.
no code implementations • CVPR 2021 • Apratim Bhattacharyya, Daniel Olmeda Reino, Mario Fritz, Bernt Schiele
In this work, we propose Euro-PVI, a dataset of pedestrian and bicyclist trajectories.
Ranked #1 on
Pedestrian Trajectory Prediction
on Euro-PVI
1 code implementation • 29 May 2021 • Yang He, Ning Yu, Margret Keuper, Mario Fritz
The rapid advances in deep generative models over the past years have led to highly {realistic media, known as deepfakes,} that are commonly indistinguishable from real to human eyes.
1 code implementation • CVPR 2021 • Moritz Böhle, Mario Fritz, Bernt Schiele
Given the alignment of the DAUs, the resulting contribution maps align with discriminative input patterns.
1 code implementation • ICCV 2021 • Ning Yu, Guilin Liu, Aysegul Dundar, Andrew Tao, Bryan Catanzaro, Larry Davis, Mario Fritz
Lastly, we study different attention architectures in the discriminator, and propose a reference attention mechanism.
no code implementations • 9 Feb 2021 • Sahar Abdelnabi, Mario Fritz
Machine learning models are now widely deployed in real-world applications.
1 code implementation • 4 Feb 2021 • Yugeng Liu, Rui Wen, Xinlei He, Ahmed Salem, Zhikun Zhang, Michael Backes, Emiliano De Cristofaro, Mario Fritz, Yang Zhang
As a result, we lack a comprehensive picture of the risks caused by the attacks, e. g., the different scenarios they can be applied to, the common factors that influence their performance, the relationship among them, or the effectiveness of possible defenses.
1 code implementation • ICLR 2022 • Ning Yu, Vladislav Skripniuk, Dingfan Chen, Larry Davis, Mario Fritz
Over the past years, deep generative models have achieved a new level of performance.
no code implementations • 15 Dec 2020 • Yang He, Hui-Po Wang, Maximilian Zenk, Mario Fritz
Despite notable progress in gradient compression, the existing quantization methods require further improvement when low-bits compression is applied, especially the overall systems often degenerate a lot when quantization are applied in double directions to compress model weights and gradients.
2 code implementations • CVPR 2021 • Hui-Po Wang, Ning Yu, Mario Fritz
While Generative Adversarial Networks (GANs) show increasing performance and the level of realism is becoming indistinguishable from natural images, this also comes with high demands on data and computation.
no code implementations • NeurIPS Workshop CAP 2020 • Hossein Hajipour, Apratim Bhattacharyya, Mario Fritz
Therefore, we propose a deep generative model to automatically correct programming errors by learning a distribution over potential fixes.
no code implementations • 21 Sep 2020 • Apratim Bhattacharyya, Christoph-Nikolas Straehle, Mario Fritz, Bernt Schiele
This yields an exact inference method that models trajectories at different spatio-temporal resolutions in a hierarchical manner.
no code implementations • 18 Sep 2020 • Yang He, Bernt Schiele, Mario Fritz
Recently, learning-based image synthesis has enabled to generate high-resolution images, either applying popular adversarial training or a powerful perceptual loss.
1 code implementation • 7 Sep 2020 • Sahar Abdelnabi, Mario Fritz
In this paper, we study natural language watermarking as a defense to help better mark and trace the provenance of text.
no code implementations • 1 Sep 2020 • Shadi Rahimian, Tribhuvanesh Orekondy, Mario Fritz
Our work consists of two sides: We introduce sampling attack, a novel membership inference technique that unlike other standard membership adversaries is able to work under severe restriction of no access to scores of the victim model.
1 code implementation • ICCV 2021 • Ning Yu, Vladislav Skripniuk, Sahar Abdelnabi, Mario Fritz
Thus, we seek a proactive and sustainable solution on deepfake detection, that is agnostic to the evolution of generative models, by introducing artificial fingerprints into the models.
no code implementations • NeurIPS Workshop CAP 2020 • Hossein Hajipour, Mateusz Malinowski, Mario Fritz
In this work, we investigate the problem of revealing the functionality of a black-box agent.
1 code implementation • NeurIPS 2020 • Dingfan Chen, Tribhuvanesh Orekondy, Mario Fritz
The wide-spread availability of rich data has fueled the growth of machine learning applications in numerous domains.
no code implementations • 20 May 2020 • Hui-Po Wang, Tribhuvanesh Orekondy, Mario Fritz
Personal photos of individuals when shared online, apart from exhibiting a myriad of memorable details, also reveals a wide range of private information and potentially entails privacy risks (e. g., online harassment, tracking).
1 code implementation • CVPR 2020 • Shweta Mahajan, Apratim Bhattacharyya, Mario Fritz, Bernt Schiele, Stefan Roth
Flow-based generative models are an important class of exact inference models that admit efficient inference and sampling for image synthesis.
1 code implementation • 7 Apr 2020 • Saurabh Sharma, Ning Yu, Mario Fritz, Bernt Schiele
Deep learning enables impressive performance in image recognition using large-scale artificially-balanced datasets.
Ranked #21 on
Long-tail Learning
on Places-LT
1 code implementation • ECCV 2020 • Ning Yu, Ke Li, Peng Zhou, Jitendra Malik, Larry Davis, Mario Fritz
Generative Adversarial Networks (GANs) have brought about rapid progress towards generating photorealistic images.
1 code implementation • 2 Mar 2020 • Tahleen Rahman, Mario Fritz, Michael Backes, Yang Zhang
Most previous works in privacy of Online Social Networks (OSN) focus on a restricted scenario of using one type of information to infer another type of information or using only static profile data such as username, profile picture or home location.
Social and Information Networks
1 code implementation • ECCV 2020 • Yang He, Shadi Rahimian, Bernt Schiele, Mario Fritz
Today's success of state of the art methods for semantic segmentation is driven by large datasets.
no code implementations • CVPR 2020 • Vedika Agarwal, Rakshith Shetty, Mario Fritz
Despite significant success in Visual Question Answering (VQA), VQA models have been shown to be notoriously brittle to linguistic variations in the questions.
1 code implementation • 27 Sep 2019 • Apratim Bhattacharyya, Mario Fritz, Bernt Schiele
Recent works have proposed hybrid VAE-GAN frameworks which integrate a GAN-based synthetic likelihood to the VAE objective to address both the mode collapse and sample quality issues, with limited success.
no code implementations • 25 Sep 2019 • Apratim Bhattacharyya, Mario Fritz, Bernt Schiele
Recent works have proposed hybrid VAE-GAN frameworks which integrate a GAN-based synthetic likelihood to the VAE objective to address both the mode collapse and sample quality issues, with limited success.
1 code implementation • 9 Sep 2019 • Dingfan Chen, Ning Yu, Yang Zhang, Mario Fritz
In addition, we propose the first generic attack model that can be instantiated in a large range of settings and is applicable to various kinds of deep generative models.
no code implementations • 1 Sep 2019 • Sahar Abdelnabi, Katharina Krombholz, Mario Fritz
Phishing websites are still a major threat in today's Internet ecosystem.
no code implementations • 24 Aug 2019 • Apratim Bhattacharyya, Michael Hanselmann, Mario Fritz, Bernt Schiele, Christoph-Nikolas Straehle
Prediction of future states of the environment and interacting agents is a key competence required for autonomous agents to operate successfully in the real world.
Ranked #12 on
Trajectory Prediction
on Stanford Drone
no code implementations • 25 Jul 2019 • Max Losch, Mario Fritz, Bernt Schiele
Additionally we show how the activations of the SB-Layer can be used for both the interpretation of failure cases of the network as well as for confidence prediction of the resulting output.
no code implementations • ICLR 2020 • Tribhuvanesh Orekondy, Bernt Schiele, Mario Fritz
We find such passive defenses ineffective against DNN stealing attacks.
no code implementations • 24 Jun 2019 • Hossein Hajipour, Apratim Bhattacharyya, Cristian-Alexandru Staicu, Mario Fritz
Therefore, we propose a generative model that learns a distribution over potential fixes.
Ranked #3 on
Program Repair
on DeepFix
no code implementations • 27 May 2019 • Hosnieh Sattar, Katharina Krombholz, Gerard Pons-Moll, Mario Fritz
Modern approaches to pose and body shape estimation have recently achieved strong performance even under challenging real-world conditions.
no code implementations • 1 Apr 2019 • Ahmed Salem, Apratim Bhattacharya, Michael Backes, Mario Fritz, Yang Zhang
As data generation is a continuous process, this leads to ML model owners updating their models frequently with newly-collected data in an online learning scenario.
no code implementations • 17 Dec 2018 • Rakshith Shetty, Bernt Schiele, Mario Fritz
We propose a method to quantify the sensitivity of black-box vision models to visual context by editing images to remove selected objects and measuring the response of the target models.
2 code implementations • CVPR 2019 • Tribhuvanesh Orekondy, Bernt Schiele, Mario Fritz
We formulate model functionality stealing as a two-step approach: (i) querying a set of input images to the blackbox model to obtain predictions; and (ii) training a "knockoff" with queried image-prediction pairs.
2 code implementations • ICCV 2019 • Ning Yu, Larry Davis, Mario Fritz
Our experiments show that (1) GANs carry distinct model fingerprints and leave stable fingerprints in their generated images, which support image attribution; (2) even minor differences in GAN training can result in different fingerprints, which enables fine-grained model authentication; (3) fingerprints persist across different image frequencies and patches and are not biased by GAN artifacts; (4) fingerprint finetuning is effective in immunizing against five types of adversarial image perturbations; and (5) comparisons also show our learned fingerprints consistently outperform several baselines in a variety of setups.
1 code implementation • ICLR 2019 • Apratim Bhattacharyya, Mario Fritz, Bernt Schiele
For autonomous agents to successfully operate in the real world, the ability to anticipate future scene states is a key competence.
1 code implementation • ECCV 2018 • Yang He, Bernt Schiele, Mario Fritz
Recent advances in Deep Learning and probabilistic modeling have led to strong improvements in generative models for images.
no code implementations • 1 Aug 2018 • Lucjan Hanzlik, Yang Zhang, Kathrin Grosse, Ahmed Salem, Max Augustin, Michael Backes, Mario Fritz
In this paper, we propose MLCapsule, a guarded offline deployment of machine learning as a service.
no code implementations • 9 Jul 2018 • Hosnieh Sattar, Gerard Pons-Moll, Mario Fritz
To study the correlation between clothing garments and body shape, we collected a new dataset (Fashion Takes Shape), which includes images of users with clothing category annotations.
1 code implementation • 20 Jun 2018 • Apratim Bhattacharyya, Bernt Schiele, Mario Fritz
For autonomous agents to successfully operate in the real world, anticipation of future events and states of their environment is a key competence.
Ranked #6 on
Human Pose Forecasting
on HumanEva-I
no code implementations • 18 Jun 2018 • Apratim Bhattacharyya, Mario Fritz, Bernt Schiele
For autonomous agents to successfully operate in the real world, anticipation of future events and states of their environment is a key competence.
no code implementations • NeurIPS 2018 • Rakshith Shetty, Mario Fritz, Bernt Schiele
While great progress has been made recently in automatic image manipulation, it has been limited to object centric images like faces or structured scene datasets.
7 code implementations • 4 Jun 2018 • Ahmed Salem, Yang Zhang, Mathias Humbert, Pascal Berrang, Mario Fritz, Michael Backes
In addition, we propose the first effective defense mechanisms against such broader class of membership inference attacks that maintain a high level of utility of the ML model.
no code implementations • CVPR 2018 • Apratim Bhattacharyya, Bernt Schiele, Mario Fritz
For autonomous agents to successfully operate in the real world, anticipation of future events and states of their environment is a key competence.
no code implementations • 31 May 2018 • Edgar Tretschk, Seong Joon Oh, Mario Fritz
As a result of our attack, the victim agent is misguided to optimise for the adversarial reward over time.
no code implementations • 15 May 2018 • Tribhuvanesh Orekondy, Seong Joon Oh, Yang Zhang, Bernt Schiele, Mario Fritz
At the core of FL is a network of anonymous user devices sharing training information (model parameter updates) computed locally on personal data.
no code implementations • ECCV 2018 • Qianru Sun, Ayush Tewari, Weipeng Xu, Mario Fritz, Christian Theobalt, Bernt Schiele
As more and more personal photos are shared and tagged in social media, avoiding privacy risks such as unintended recognition becomes increasingly challenging.
no code implementations • ICCV 2019 • Maxim Maximov, Laura Leal-Taixé, Mario Fritz, Tobias Ritschel
Second, we demonstrate how another network can be used to map from an image or video frames to a DAM network to reproduce this appearance, without using a lengthy optimization such as stochastic gradient descent (learning-to-learn).
1 code implementation • CVPR 2018 • Liqian Ma, Qianru Sun, Stamatios Georgoulis, Luc van Gool, Bernt Schiele, Mario Fritz
Generating novel, yet realistic, images of persons is a challenging task due to the complex interplay between the different image factors, such as the foreground, background and pose information.
Ranked #2 on
Gesture-to-Gesture Translation
on Senz3D
no code implementations • CVPR 2018 • Tribhuvanesh Orekondy, Mario Fritz, Bernt Schiele
Images convey a broad spectrum of personal information.
no code implementations • CVPR 2018 • Apratim Bhattacharyya, Mario Fritz, Bernt Schiele
Our experimental results show that it is indeed possible to predict people trajectories at the desired time horizons and that our uncertainty estimates are informative of the prediction error.
Ranked #5 on
Trajectory Prediction
on JAAD
6 code implementations • 24 Nov 2017 • Xucong Zhang, Yusuke Sugano, Mario Fritz, Andreas Bulling
Second, we present an extensive evaluation of state-of-the-art gaze estimation methods on three current datasets, including MPIIGaze.
no code implementations • CVPR 2018 • Qianru Sun, Liqian Ma, Seong Joon Oh, Luc van Gool, Bernt Schiele, Mario Fritz
As more and more personal photos are shared online, being able to obfuscate identities in such photos is becoming a necessity for privacy protection.
no code implementations • 6 Nov 2017 • Rakshith Shetty, Bernt Schiele, Mario Fritz
In this paper, we propose an automatic method, called Adversarial Author Attribute Anonymity Neural Translation ($A^4NT$), to combat such text-based adversaries.
3 code implementations • ICLR 2018 • Seong Joon Oh, Max Augustin, Bernt Schiele, Mario Fritz
On the one hand, our work exposes the vulnerability of black-box neural networks to different types of attacks -- we show that the revealed internal information helps generate more effective adversarial examples against the black box model.
no code implementations • ICLR 2018 • Wenbin Li, Jeannette Bohg, Mario Fritz
We created a synthetic block stacking environment with physics simulation in which the agent can learn a policy end-to-end through trial and error.
no code implementations • 9 Oct 2017 • Seong Joon Oh, Rodrigo Benenson, Mario Fritz, Bernt Schiele
Person recognition in social media photos sets new challenges for computer vision, including non-cooperative subjects (e. g. backward viewpoints, unusual poses) and great changes in appearance.
1 code implementation • 6 Sep 2017 • Yang He, Margret Keuper, Bernt Schiele, Mario Fritz
In this paper, we present an approach for learning dilation parameters adaptively per channel, consistently improving semantic segmentation results on street-scene datasets like Cityscapes and Camvid.
no code implementations • 20 Jun 2017 • Seyed Majid Azimi, Dominik Britz, Michael Engstler, Mario Fritz, Frank Mücklich
In this work, we propose a Deep Learning method for microstructural classification in the examples of certain microstructural constituents of low carbon steel.
no code implementations • 19 Jun 2017 • Hosnieh Sattar, Mario Fritz, Andreas Bulling
Such visual decoding is challenging for two reasons: 1) the search target only resides in the user's mind as a subjective visual pattern, and can most often not even be described verbally by the person, and 2) it is, as of yet, unclear if gaze fixations contain sufficient information for this task at all.
1 code implementation • CVPR 2017 • Qianru Sun, Bernt Schiele, Mario Fritz
Social relations are the foundation of human daily life.
Ranked #6 on
Visual Social Relationship Recognition
on PIPA
no code implementations • ICCV 2017 • Tribhuvanesh Orekondy, Bernt Schiele, Mario Fritz
Third, we propose models that predict user specific privacy score from images in order to enforce the users' privacy preferences.
3 code implementations • ICCV 2017 • Rakshith Shetty, Marcus Rohrbach, Lisa Anne Hendricks, Mario Fritz, Bernt Schiele
While strong progress has been made in image captioning over the last years, machine and human captions are still quite distinct.
no code implementations • ICCV 2017 • Seong Joon Oh, Mario Fritz, Bernt Schiele
We derive the optimal strategy for the user that assures an upper bound on the recognition rate independent of the recogniser's counter measure.
no code implementations • CVPR 2017 • Seong Joon Oh, Rodrigo Benenson, Anna Khoreva, Zeynep Akata, Mario Fritz, Bernt Schiele
We show how to combine both information sources in order to recover 80% of the fully supervised performance - which is the new state of the art in weakly supervised training for pixel-wise semantic labelling.
Ranked #26 on
Semantic Segmentation
on PASCAL VOC 2012 val
no code implementations • ICCV 2017 • Stamatios Georgoulis, Konstantinos Rematas, Tobias Ritschel, Mario Fritz, Tinne Tuytelaars, Luc van Gool
How much does a single image reveal about the environment it was taken in?
no code implementations • 27 Nov 2016 • Hosnieh Sattar, Andreas Bulling, Mario Fritz
Predicting the target of visual search from eye fixation (gaze) data is a challenging problem with many applications in human-computer interaction.
no code implementations • 27 Nov 2016 • Apratim Bhattacharyya, Mateusz Malinowski, Bernt Schiele, Mario Fritz
Boundary estimation in images and videos has been a very active topic of research, and organizing visual information into boundaries and segments is believed to be a corner stone of visual perception.
4 code implementations • 27 Nov 2016 • Xucong Zhang, Yusuke Sugano, Mario Fritz, Andreas Bulling
Eye gaze is an important non-verbal cue for human affect analysis.
1 code implementation • 4 Oct 2016 • Mateusz Malinowski, Mario Fritz
Together with the development of more accurate methods in Computer Vision and Natural Language Understanding, holistic architectures that answer on questions about the content of real-world images have emerged.
no code implementations • 15 Sep 2016 • Wenbin Li, Aleš Leonardis, Mario Fritz
We present a learning-based approach based on simulated data that predicts stability of towers comprised of wooden blocks under different conditions and quantities related to the potential fall of the towers.
no code implementations • 3 Sep 2016 • Wei-Chen Chiu, Fabio Galasso, Mario Fritz
Are we ready to segment consumer stereo videos?
no code implementations • 9 Aug 2016 • Ashkan Mokarian, Mateusz Malinowski, Mario Fritz
We present Mean Box Pooling, a novel visual representation that pools over CNN representations of a large number, highly overlapping object proposals.
no code implementations • 28 Jul 2016 • Seong Joon Oh, Rodrigo Benenson, Mario Fritz, Bernt Schiele
As we shift more of our lives into the virtual domain, the volume of data shared on the web keeps increasing and presents a threat to our privacy.
no code implementations • 24 May 2016 • Apratim Bhattacharyya, Mateusz Malinowski, Mario Fritz
Furthermore, we show long-term prediction of boundaries in situations where the motion is governed by the laws of physics.
1 code implementation • 9 May 2016 • Mateusz Malinowski, Marcus Rohrbach, Mario Fritz
By combining latest advances in image representation and natural language processing, we propose Ask Your Neurons, a scalable, jointly trained, end-to-end formulation to this problem.
1 code implementation • 13 Apr 2016 • Abhishek Sharma, Oliver Grau, Mario Fritz
Prior work has shown encouraging results on problems ranging from shape completion to recognition.
1 code implementation • CVPR 2017 • Yang He, Wei-Chen Chiu, Margret Keuper, Mario Fritz
The proposed network produces a high quality segmentation of a single image by leveraging information from additional views of the same scene.
Ranked #110 on
Semantic Segmentation
on NYU Depth v2
no code implementations • 31 Mar 2016 • Wenbin Li, Seyedmajid Azimi, Aleš Leonardis, Mario Fritz
In this paper, we contrast a more traditional approach of taking a model-based route with explicit 3D representations and physical simulation by an end-to-end approach that directly predicts stability and related quantities from appearance.
no code implementations • CVPR 2016 • Zeynep Akata, Mateusz Malinowski, Mario Fritz, Bernt Schiele
A promising research direction is zero-shot learning, which does not require any training data to recognize new classes, but rather relies on some form of auxiliary information describing the new classes.
no code implementations • 27 Mar 2016 • Stamatios Georgoulis, Konstantinos Rematas, Tobias Ritschel, Mario Fritz, Luc van Gool, Tinne Tuytelaars
In this paper we are extracting surface reflectance and natural environmental illumination from a reflectance map, i. e. from a single 2D image of a sphere of one material under one illumination.
no code implementations • 16 Feb 2016 • Sreyasi Nag Chowdhury, Mateusz Malinowski, Andreas Bulling, Mario Fritz
We show that our retrieval system can cope with this variability using personalisation through an online learning-based retrieval formulation.
no code implementations • 31 Jan 2016 • Konstantinos Rematas, Chuong Nguyen, Tobias Ritschel, Mario Fritz, Tinne Tuytelaars
We propose a technique to use the structural information extracted from a 3D model that matches the image object in terms of viewpoint and shape.
no code implementations • CVPR 2016 • Konstantinos Rematas, Tobias Ritschel, Mario Fritz, Efstratios Gavves, Tinne Tuytelaars
Undoing the image formation process and therefore decomposing appearance into its intrinsic properties is a challenging task due to the under-constraint nature of this inverse problem.
no code implementations • ICCV 2015 • Seong Joon Oh, Rodrigo Benenson, Mario Fritz, Bernt Schiele
Recognising persons in everyday photos presents major challenges (occluded faces, different clothing, locations, etc.)
no code implementations • 21 May 2015 • Iaroslav Shcherbatyi, Andreas Bulling, Mario Fritz
An increasing number of works explore collaborative human-computer systems in which human gaze is used to enhance computer vision systems.
no code implementations • ICCV 2015 • Mateusz Malinowski, Marcus Rohrbach, Mario Fritz
In contrast to previous efforts, we are facing a multi-modal problem where the language output (answer) is conditioned on visual and natural language input (image and question).
no code implementations • ICCV 2015 • Wei-Chen Chiu, Mario Fritz
The Histogram of Oriented Gradient (HOG) descriptor has led to many advances in computer vision over the last decade and is still part of many state of the art approaches.
6 code implementations • CVPR 2015 • Xucong Zhang, Yusuke Sugano, Mario Fritz, Andreas Bulling
Appearance-based gaze estimation is believed to work well in real-world settings, but existing datasets have been collected under controlled laboratory conditions and methods have been not evaluated across multiple datasets.
no code implementations • CVPR 2015 • Hosnieh Sattar, Sabine Müller, Mario Fritz, Andreas Bulling
Previous work on predicting the target of visual search from human fixations only considered closed-world settings in which training labels are available and predictions are performed for a known set of potential targets.
no code implementations • 14 Jan 2015 • Mateusz Malinowski, Mario Fritz
Progress in language and image understanding by machines has sparkled the interest of the research community in more open-ended, holistic tasks, and refueled an old AI dream of building intelligent machines.
no code implementations • 19 Nov 2014 • Mateusz Malinowski, Mario Fritz
Over the last two decades we have witnessed strong progress on modeling visual object classes, scenes and attributes that have significantly contributed to automated image understanding.
no code implementations • 29 Oct 2014 • Mateusz Malinowski, Mario Fritz
As language and visual understanding by machines progresses rapidly, we are observing an increasing interest in holistic architectures that tightly interlink both modalities in a joint learning and inference process.
no code implementations • NeurIPS 2014 • Mateusz Malinowski, Mario Fritz
We propose a method for automatically answering questions about images by bringing together recent advances from natural language processing and computer vision.
no code implementations • 13 Aug 2014 • Wenbin Li, Mario Fritz
The recent progress in sparse coding and deep learning has made unsupervised feature learning methods a strong competitor to hand-crafted descriptors.
no code implementations • CVPR 2014 • Sergey Karayev, Mario Fritz, Trevor Darrell
On suitable datasets, we can incorporate a semantic back-off strategy that gives maximally specific predictions for a desired level of accuracy; this provides a new view on the time course of human visual perception.
no code implementations • CVPR 2014 • Konstantinos Rematas, Tobias Ritschel, Mario Fritz, Tinne Tuytelaars
We propose a technique to use the structural information extracted from a set of 3D models of an object class to improve novel-view synthesis for images showing unknown instances of this class.
no code implementations • 6 Mar 2014 • Mark Simkin, Dominique Schroeder, Andreas Bulling, Mario Fritz
We describe Ubic, a framework that allows users to bridge the gap between digital cryptography and the physical world.
no code implementations • CVPR 2013 • Wei-Chen Chiu, Mario Fritz
This is a clear mismatch to the challenges that we are facing with videos from online resources or consumer videos.
no code implementations • 15 Jan 2013 • Mateusz Malinowski, Mario Fritz
Biologically inspired, from the early HMAX model to Spatial Pyramid Matching, pooling has played an important role in visual recognition pipelines.
no code implementations • NeurIPS 2012 • Sergey Karayev, Tobias Baumgartner, Mario Fritz, Trevor Darrell
On the timeliness measure, our method obtains at least $11\%$ better performance.
no code implementations • NeurIPS 2010 • Mario Fritz, Kate Saenko, Trevor Darrell
Metric constraints are known to be highly discriminative for many objects, but if training is limited to data captured from a particular 3-D sensor the quantity of training data may be severly limited.
no code implementations • NeurIPS 2009 • Mario Fritz, Gary Bradski, Sergey Karayev, Trevor Darrell, Michael J. Black
The appearance of a transparent patch is determined in part by the refraction of a background pattern through a transparent medium: the energy from the background usually dominates the patch appearance.