Patrick Aichroth

Fraunhofer Institute for Digital Media Technology

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Trends

Author Order

Document Type

Co-Authors

Luca Cuccovillo

Fraunhofer Institute for Digital Media Technology

Sebastian Mann

Friedrich-Alexander-Universität Erlangen-Nürnberg

Milica Gerhardt

Fraunhofer Institute for Digital Media Technology

Stefano Tubaro

Politecnico di Milano

Thomas Köllmer

Fraunhofer Institute for Digital Media Technology

Artem Yaroshchuk

Fraunhofer Institute for Digital Media Technology

Christian Dittmar

Fraunhofer Institute for Integrated Circuits

Paolo Bestagini

Politecnico di Milano

Milica Maksimović

Fraunhofer Institute for Digital Media Technology

Marco Tagliasacchi

Google (Switzerland)

Cooperative Institutions

Fraunhofer Institute for Digital Media Technology

Politecnico di Milano

Fraunhofer Society

Technische Universität Ilmenau

Fraunhofer Institute for Integrated Circuits

Consorzio di Bioingegneria e Informatica Medica

Technical University of Munich

Information Technologies Institute

Centre for Research and Technology Hellas

Fraunhofer Institute for Experimental Software Engineering

Author Statistics

Papers

Citation

H-Index

i-10 index

Research Field

Spectral Denoising for Microphone Classification

Luca Cuccovillo Antonio Giganti Paolo Bestagini Patrick Aichroth Stefano Tubaro

In this paper, we propose the use of denoising for microphone classification, to enable its usage for several key application domains that involve noisy conditions. We describe the proposed analysis pipeline and the baseline algorithm for microphone classification, and discuss various denoising approaches which can be applied to it within the time or spectral domain; finally, we determine the best-performing denoising procedure, and evaluate the performance of the overall, integrated approach with several SNR levels of additive input noise. As a result, the proposed method achieves an average accuracy increase of about 25% on denoised content over the reference baseline.

Video denoising

Baseline (sea)

10.1145/3512732.3533586

Cite

Citations (4)

Video camera identification using audio-visual features

Simone Milani Luca Cuccovillo Marco Tagliasacchi Stefano Tubaro Patrick Aichroth

One of the major issues in multimedia forensics is the identification of video acquisition devices. Most of the relevant state-of-the-art solutions rely on either visual or audio analysis, using feature arrays that are highly correlated with the characteristics of the respective camera or microphone. In this work, we present a multi-modal approach that uses both video and audio information to improve the detection accuracy. For this purpose, microphone detection based on the blind estimation of the frequency response is complemented with a video camera detection based on a set of video features related to the Color Filter Array interpolation. Experimental results show that the combined approach results in an improved overall classification accuracy over the mono-modal cases.

Audio analyzer

Feature (linguistics)

Identification

Interpolation

Audio visual

10.1109/euvip.2014.7018382

Cite

Citations (4)

Freebies for CD Retailers : A Transition from Tangible to Digital Distribution

Jens. Hasselbach Patrick Aichroth

10.1400/56818

Cite

Citations (0)

Environment Classification via Blind Roomprints Estimation

arXiv (Cornell University) (2022)

Malte Baum Luca Cuccovillo Artem Yaroshchuk Patrick Aichroth

In this paper we present a novel approach for environment classification for speech recordings, which does not require the selection of decaying reverberation tails. It is based on a multi-band RT60 analysis of blind channel estimates and achieves an accuracy of up to 93.6% on test recordings derived from the ACE corpus.

10.48550/arxiv.2209.07196

Cite

Citations (0)

SEC-Learn: Sensor Edge Cloud for Federated Learning

Lecture notes in computer science (2022)

Patrick Aichroth Christoph Antes Pierre Gembatzka Holger Gräf David S. Johnson

Neuromorphic engineering

Edge device

10.1007/978-3-031-04580-6_29

Cite

Citations (0)

The MICO Broker: An Orchestration Framework for Linked Data Extractors.

Patrick Aichroth Marcel Sieland Luca Cuccovillo Thomas Köllmer

Orchestration

Source

Cite

Citations (1)

Audio forensics and provenance analysis: Technologies for media verification and asset management

Journal of digital media management (2021)

Patrick Aichroth Luca Cuccovillo Milica Gerhardt

This paper discusses a number of audio forensics and provenance analysis technologies that were developed for media verification and media asset management, describing the motivation behind them, the methods used to develop them, and the way they are applied in practice. The paper also shows how the technologies may be adapted and applied to other domains. Finally, it outlines why it is crucial for research and development to follow an iterative, interactive process that does not require large budgets, but rather the willingness of users (here: broadcasters) and researchers to conduct experiments with limited risk and investment instead of waiting for ‘perfect projects’ and ‘perfect technologies’.

10.69554/cexg6223

Cite

Citations (3)

A Workflow for Cross Media Recommendations based on Linked Data Analysis.

Thomas Köllmer Emanuel Berndl Thomas Weißgerber Patrick Aichroth Harald Kosch

Linked Data

Source

Cite

Citations (2)

Advancing Audio Phylogeny: A Neural Network Approach for Transformation Detection

Milica Gerhardt Luca Cuccovillo Patrick Aichroth

In this study we propose a novel approach to audio phylogeny, i.e. the detection of relationships and transformations within a set of near-duplicate audio items, by leveraging a deep neural network for efficiency and extensibility. Unlike existing methods, our approach detects transformations between nodes in one step, and the transformation set can be expanded by retraining the neural network without excessive computational costs. We evaluated our method against the state of the art using a self-created and publicly released dataset, observing a superior performance in reconstructing phylogenetic trees and heightened transformation detection accuracy. Moreover, the ability to detect a wide range of transformations and to extend the transformation set make the approach suitable for various applications.

Deep Neural Networks

10.1109/wifs58808.2023.10375058

Cite

Citations (1)

Open Challenges in Synthetic Speech Detection

arXiv (Cornell University) (2022)

Luca Cuccovillo Christoforos Papastergiopoulos Anastasios Vafeiadis Artem Yaroshchuk Patrick Aichroth

In this paper the current status and open challenges of synthetic speech detection are addressed. The work comprises an initial analysis of available open datasets and of existing detection methods, a description of the requirements for new research datasets compliant with regulations and better representing real-case scenarios, and a discussion of the desired characteristics of future trustworthy detection methods in terms of both functional and non-functional requirements. Compared to other works, based on specific detection solutions or presenting single dataset of synthetic speeches, our paper is meant to orient future state-of-the-art research in the domain, to quickly lessen the current gap between synthesis and detection approaches.

Open research

Open domain

Trustworthiness

10.48550/arxiv.2209.07180

Cite

Citations (0)