Search Results for author: Andreas Dengel

Found 129 papers, 46 papers with code

Explainability of Sub-Field Level Crop Yield Prediction using Remote Sensing

no code implementations11 Jul 2024 Hiba Najjar, Miro Miranda, Marlon Nuske, Ribana Roscher, Andreas Dengel

Our goal is to develop and explain predictive models for these crops, using a large dataset of satellite images, additional data modalities, and crop yield maps.

Crop Yield Prediction Decision Making +1

DocXplain: A Novel Model-Agnostic Explainability Method for Document Image Classification

no code implementations4 Jul 2024 Saifullah Saifullah, Stefan Agne, Andreas Dengel, Sheraz Ahmed

In this paper, we aim to bridge this research gap by introducing DocXplain, a novel model-agnostic explainability method specifically designed for generating high interpretability feature attribution maps for the task of document image classification.

Document Image Classification Fairness +1

VAEneu: A New Avenue for VAE Application on Probabilistic Forecasting

no code implementations7 May 2024 Alireza Koochali, Ensiye Tahaei, Andreas Dengel, Sheraz Ahmed

This paper presents VAEneu, an innovative autoregressive method for multistep ahead univariate probabilistic time series forecasting.

Probabilistic Time Series Forecasting Time Series

Federated Learning for Blind Image Super-Resolution

no code implementations26 Apr 2024 Brian B. Moser, Ahmed Anwar, Federico Raue, Stanislav Frolov, Andreas Dengel

This distinction is crucial as it circumvents the need to precisely model real-world degradations, which limits contemporary blind image SR research.

Federated Learning Image Super-Resolution

Fin-Fed-OD: Federated Outlier Detection on Financial Tabular Data

no code implementations23 Apr 2024 Dayananda Herurkar, Sebastian Palacio, Ahmed Anwar, Joern Hees, Andreas Dengel

Anomaly detection in real-world scenarios poses challenges due to dynamic and often unknown anomaly distributions, requiring robust methods that operate under an open-world assumption.

Anomaly Detection Federated Learning +2

Data Collection of Real-Life Knowledge Work in Context: The RLKWiC Dataset

no code implementations16 Apr 2024 Mahta Bakhshizadeh, Christian Jilek, Markus Schröder, Heiko Maus, Andreas Dengel

Over the years, various approaches have been employed to enhance the productivity of knowledge workers, from addressing psychological well-being to the development of personal knowledge assistants.

Benchmarking Management +1

ObjBlur: A Curriculum Learning Approach With Progressive Object-Level Blurring for Improved Layout-to-Image Generation

no code implementations11 Apr 2024 Stanislav Frolov, Brian B. Moser, Sebastian Palacio, Andreas Dengel

We present ObjBlur, a novel curriculum learning approach to improve layout-to-image generation models, where the task is to produce realistic images from layouts composed of boxes and labels.

Layout-to-Image Generation

A Study in Dataset Pruning for Image Super-Resolution

no code implementations25 Mar 2024 Brian B. Moser, Federico Raue, Andreas Dengel

By focusing the training on just 50\% of the original dataset, specifically on the samples characterized by the highest loss values, we achieve results comparable to or surpassing those obtained from training on the entire dataset.

Image Super-Resolution

Impact Assessment of Missing Data in Model Predictions for Earth Observation Applications

1 code implementation21 Mar 2024 Francisco Mena, Diego Arenas, Marcela Charfuelan, Marlon Nuske, Andreas Dengel

In this work, we assess the impact of missing temporal and static EO sources in trained models across four datasets with classification and regression tasks.

Crop Classification Earth Observation +1

Latent Dataset Distillation with Diffusion Models

no code implementations6 Mar 2024 Brian B. Moser, Federico Raue, Sebastian Palacio, Stanislav Frolov, Andreas Dengel

In response to these limitations, the concept of distilling the information on a dataset into a condensed set of (synthetic) samples, namely a distilled dataset, emerged.

Dataset Distillation

TaylorShift: Shifting the Complexity of Self-Attention from Squared to Linear (and Back) using Taylor-Softmax

1 code implementation5 Mar 2024 Tobias Christian Nauen, Sebastian Palacio, Andreas Dengel

The quadratic complexity of the attention mechanism represents one of the biggest hurdles for processing long sequences using Transformers.

Classification

Opening the Black-Box: A Systematic Review on Explainable AI in Remote Sensing

no code implementations21 Feb 2024 Adrian Höhl, Ivica Obadic, Miguel Ángel Fernández Torres, Hiba Najjar, Dario Oliveira, Zeynep Akata, Andreas Dengel, Xiao Xiang Zhu

In recent years, black-box machine learning approaches have become a dominant modeling paradigm for knowledge extraction in Remote Sensing.

Modular Graph Extraction for Handwritten Circuit Diagram Images

no code implementations16 Feb 2024 Johannes Bayer, Leo van Waveren, Andreas Dengel

However, apart from printed legacy schematics, hand-drawn circuit diagrams are still used today in the educational domain, where they serve as an easily accessible mean for trainees and students to learn drawing this type of diagrams.

object-detection Object Detection

Real-time Traffic Object Detection for Autonomous Driving

no code implementations31 Jan 2024 Abdul Hannan Khan, Syed Tahseen Raza Rizvi, Andreas Dengel

In this research, we assess the robustness of our previously proposed, highly efficient pedestrian detector LSFM on well-established autonomous driving benchmarks, including diverse weather conditions and nighttime scenes.

Autonomous Driving Object +3

Diffusion Models, Image Super-Resolution And Everything: A Survey

no code implementations1 Jan 2024 Brian B. Moser, Arundhati S. Shanbhag, Federico Raue, Stanislav Frolov, Sebastian Palacio, Andreas Dengel

Diffusion Models (DMs) have disrupted the image Super-Resolution (SR) field and further closed the gap between image quality and human perceptual preferences.

Computational Efficiency Image Super-Resolution +1

Class Conditional Time Series Generation with Structured Noise Space GAN

no code implementations20 Dec 2023 Hamidreza Gholamrezaei, Alireza Koochali, Andreas Dengel, Sheraz Ahmed

This paper introduces Structured Noise Space GAN (SNS-GAN), a novel approach in the field of generative modeling specifically tailored for class-conditional generation in both image and time series data.

Time Series Time Series Generation

Q-Seg: Quantum Annealing-based Unsupervised Image Segmentation

1 code implementation21 Nov 2023 Supreeth Mysore Venkatesh, Antonio Macaluso, Marlon Nuske, Matthias Klusch, Andreas Dengel

Thus, Q-Seg emerges as a viable alternative for real-world applications using available quantum hardware, particularly in scenarios where the lack of labeled data and computational runtime are critical.

Earth Observation Image Segmentation +3

Medi-CAT: Contrastive Adversarial Training for Medical Image Classification

no code implementations31 Oct 2023 Pervaiz Iqbal Khan, Andreas Dengel, Sheraz Ahmed

This paper proposes a training strategy Medi-CAT to overcome the underfitting and overfitting phenomena in medical imaging datasets.

Contrastive Learning Image Classification +1

A Unique Training Strategy to Enhance Language Models Capabilities for Health Mention Detection from Social Media Content

no code implementations29 Oct 2023 Pervaiz Iqbal Khan, Muhammad Nabeel Asim, Andreas Dengel, Sheraz Ahmed

Following the need for an optimal language model competent in extracting useful patterns from social media text, the key goal of this paper is to train language models in such a way that they learn to derive generalized patterns.

Contrastive Learning Language Modelling

PrIeD-KIE: Towards Privacy Preserved Document Key Information Extraction

no code implementations5 Oct 2023 Saifullah Saifullah, Stefan Agne, Andreas Dengel, Sheraz Ahmed

We conduct a comprehensive evaluation of the algorithm across various client and privacy settings, and demonstrate its capability to achieve comparable performance and privacy guarantees to standalone DP, even when accommodating an increasing number of participating clients.

Document AI Federated Learning +1

Segment Anything for Microscopy

1 code implementation bioRxiv 2023 Anwai Archit, Sushmita Nair, Nabeel Khalid, Paul Hilt, Vikas Rajashekar, Marei Freitag, Sagnik Gupta, Andreas Dengel, Sheraz Ahmed, Constantin Pape

We present Segment Anything for Microscopy, a tool for interactive and automatic segmentation and tracking of objects in multi-dimensional microscopy data.

Image Segmentation Instance Segmentation +3

Which Transformer to Favor: A Comparative Analysis of Efficiency in Vision Transformers

1 code implementation18 Aug 2023 Tobias Christian Nauen, Sebastian Palacio, Andreas Dengel

This benchmark provides a standardized baseline across the landscape of efficiency-oriented transformers and our framework of analysis, based on Pareto optimality, reveals surprising insights.

Image Classification Model Selection

Dynamic Attention-Guided Diffusion for Image Super-Resolution

no code implementations15 Aug 2023 Brian B. Moser, Stanislav Frolov, Federico Raue, Sebastian Palacio, Andreas Dengel

To address this, we introduce "You Only Diffuse Areas" (YODA), a dynamic attention-guided diffusion method for image SR. YODA selectively focuses on spatial regions using attention maps derived from the low-resolution image and the current time step in the diffusion process.

Image Super-Resolution SSIM

A Comparative Assessment of Multi-view fusion learning for Crop Classification

3 code implementations10 Aug 2023 Francisco Mena, Diego Arenas, Marlon Nuske, Andreas Dengel

Instead, we present a comparison of multi-view fusion methods for three different datasets and show that, depending on the test region, different methods obtain the best performance.

Crop Classification Diversity +1

Towards Self-organizing Personal Knowledge Assistants in Evolving Corporate Memories

no code implementations3 Aug 2023 Christian Jilek, Markus Schröder, Heiko Maus, Sven Schwarz, Andreas Dengel

This paper presents a retrospective overview of a decade of research in our department towards self-organizing personal knowledge assistants in evolving corporate memories.

graph construction Management

DWA: Differential Wavelet Amplifier for Image Super-Resolution

1 code implementation10 Jul 2023 Brian B. Moser, Stanislav Frolov, Federico Raue, Sebastian Palacio, Andreas Dengel

This work introduces Differential Wavelet Amplifier (DWA), a drop-in module for wavelet-based image Super-Resolution (SR).

Image Super-Resolution

DartsReNet: Exploring new RNN cells in ReNet architectures

1 code implementation11 Apr 2023 Brian Moser, Federico Raue, Jörn Hees, Andreas Dengel

We present new Recurrent Neural Network (RNN) cells for image classification using a Neural Architecture Search (NAS) approach called DARTS.

Image Classification Neural Architecture Search

From Private to Public: Benchmarking GANs in the Context of Private Time Series Classification

no code implementations28 Mar 2023 Dominique Mercier, Andreas Dengel, Sheraz Ahmed

In this work, two very prominent GAN-based architectures were evaluated in the context of private time series classification.

Benchmarking Privacy Preserving +3

A Bayesian Generative Adversarial Network (GAN) to Generate Synthetic Time-Series Data, Application in Combined Sewer Flow Prediction

no code implementations31 Jan 2023 Amin E. Bakhshipour, Alireza Koochali, Ulrich Dittmer, Ali Haghighi, Sheraz Ahmad, Andreas Dengel

In this study, we developed a GAN model to generate synthetic time series to balance our limited recorded time series data and improve the accuracy of a data-driven model for combined sewer flow prediction.

Data Augmentation Generative Adversarial Network +3

Instance Segmentation Based Graph Extraction for Handwritten Circuit Diagram Images

no code implementations9 Jan 2023 Johannes Bayer, Amit Kumar Roy, Andreas Dengel

This paper describes an approach for extracting both the electrical components (including their terminals and describing texts) as well their interconnections (including junctions and wire hops) by the means of instance segmentation and keypoint extraction.

Instance Segmentation Semantic Segmentation

Localized Semantic Feature Mixers for Efficient Pedestrian Detection in Autonomous Driving

no code implementations CVPR 2023 Abdul Hannan Khan, Mohammed Shariq Nawaz, Andreas Dengel

Autonomous driving systems rely heavily on the underlying perception module which needs to be both performant and efficient to allow precise decisions in real-time.

Autonomous Driving Pedestrian Detection

Common Practices and Taxonomy in Deep Multi-view Fusion for Remote Sensing Applications

1 code implementation20 Dec 2022 Francisco Mena, Diego Arenas, Marlon Nuske, Andreas Dengel

However, the approaches in the literature vary greatly since different terminology is used to refer to similar concepts or different illustrations are given to similar techniques.

Earth Observation

Revisiting the Shape-Bias of Deep Learning for Dermoscopic Skin Lesion Classification

1 code implementation13 Jun 2022 Adriano Lucieri, Fabian Schmeisser, Christoph Peter Balada, Shoaib Ahmed Siddiqui, Andreas Dengel, Sheraz Ahmed

Interestingly, despite deep feature extractors being inclined towards learning entangled features for skin lesion classification, individual features can still be decoded from this entangled representation.

Classification Decision Making +2

A Novel Approach to Train Diverse Types of Language Models for Health Mention Classification of Tweets

no code implementations13 Apr 2022 Pervaiz Iqbal Khan, Imran Razzak, Andreas Dengel, Sheraz Ahmed

Moreover, our analysis shows that adding noise at earlier layers improves models' performance whereas adding noise at intermediate layers deteriorates models' performance.

FiN: A Smart Grid and Power Line Communication Dataset

no code implementations13 Apr 2022 Christoph Balada, Sheraz Ahmed, Andreas Dengel, Max Bondorf, Nikolai Hopfer, Markus Zdrallek

To overcome this, power line communication (PLC) has emerged as a potential solution for reliable monitoring of the low-voltage grid.

DT2I: Dense Text-to-Image Generation from Region Descriptions

no code implementations5 Apr 2022 Stanislav Frolov, Prateek Bansal, Jörn Hees, Andreas Dengel

Our results demonstrate the capability of our approach to generate plausible images of complex scenes using region captions.

Conditional Image Generation Image-text matching +2

DocXClassifier: High Performance Explainable Deep Network for Document Image Classification

1 code implementation TechArXiv 2022 Saifullah, Stefan Agne, Andreas Dengel, Sheraz Ahmed

Our approach achieves a new peak performance in image-based classification on two popular document datasets, namely RVL-CDIP and Tobacco3482, with a top-1 classification accuracy of 94. 17% and 95. 57% on the two datasets, respectively.

Classification Data Augmentation +4

Less is More: Proxy Datasets in NAS approaches

1 code implementation14 Mar 2022 Brian Moser, Federico Raue, Jörn Hees, Andreas Dengel

One of our surprising findings is that in most cases we can reduce the amount of training data to 25\%, consequently reducing search time to 25\%, while at the same time maintaining the same accuracy as if training on the full dataset.

Neural Architecture Search

F2DNet: Fast Focal Detection Network for Pedestrian Detection

2 code implementations4 Mar 2022 Abdul Hannan Khan, Mohsin Munir, Ludger van Elst, Andreas Dengel

However, the current two-stage detectors are inefficient as they do bounding box regression in multiple steps i. e. in region proposal networks and bounding box heads.

Ranked #2 on Pedestrian Detection on Caltech (using extra training data)

object-detection Object Detection +2

Improving Health Mentioning Classification of Tweets using Contrastive Adversarial Training

no code implementations3 Mar 2022 Pervaiz Iqbal Khan, Shoaib Ahmed Siddiqui, Imran Razzak, Andreas Dengel, Sheraz Ahmed

The idea is to learn word representation by its surrounding words and utilize emojis in the text to help improve the classification results.

Utilizing Out-Domain Datasets to Enhance Multi-Task Citation Analysis

no code implementations22 Feb 2022 Dominique Mercier, Syed Tahseen Raza Rizvi, Vikas Rajashekar, Sheraz Ahmed, Andreas Dengel

However, qualitative aspects provide deeper insights into the impact of a scientific research artifact and make it possible to focus on relevant literature free from bias associated with quantitative aspects.

Scheduling Sentiment Analysis

Time to Focus: A Comprehensive Benchmark Using Time Series Attribution Methods

no code implementations8 Feb 2022 Dominique Mercier, Jwalin Bhatt, Andreas Dengel, Sheraz Ahmed

However, due to the lack of transparency the use of these networks is hampered in the areas with safety critical areas.

Time Series Time Series Analysis

KENN: Enhancing Deep Neural Networks by Leveraging Knowledge for Time Series Forecasting

no code implementations8 Feb 2022 Muhammad Ali Chattha, Ludger van Elst, Muhammad Imran Malik, Andreas Dengel, Sheraz Ahmed

End-to-end data-driven machine learning methods often have exuberant requirements in terms of quality and quantity of training data which are often impractical to fulfill in real-world applications.

Anomaly Detection Time Series +1

Random Noise vs State-of-the-Art Probabilistic Forecasting Methods : A Case Study on CRPS-Sum Discrimination Ability

no code implementations21 Jan 2022 Alireza Koochali, Peter Schichtel, Andreas Dengel, Sheraz Ahmed

The recent developments in the machine learning domain have enabled the development of complex multivariate probabilistic forecasting models.

ExAID: A Multimodal Explanation Framework for Computer-Aided Diagnosis of Skin Lesions

no code implementations4 Jan 2022 Adriano Lucieri, Muhammad Naseer Bajwa, Stephan Alexander Braun, Muhammad Imran Malik, Andreas Dengel, Sheraz Ahmed

This work presents ExAID (Explainable AI for Dermatology), a novel framework for biomedical image analysis, providing multi-modal concept-based explanations consisting of easy-to-understand textual explanations supplemented by visual maps justifying the predictions.

Decision Making

Search and Learn: Improving Semantic Coverage for Data-to-Text Generation

1 code implementation6 Dec 2021 Shailza Jolly, Zi Xuan Zhang, Andreas Dengel, Lili Mou

To this end, we propose a search-and-learning approach that leverages pretrained language models but inserts the missing slots to improve the semantic coverage.

Data-to-Text Generation

Evaluating Privacy-Preserving Machine Learning in Critical Infrastructures: A Case Study on Time-Series Classification

1 code implementation29 Nov 2021 Dominique Mercier, Adriano Lucieri, Mohsin Munir, Andreas Dengel, Sheraz Ahmed

With the advent of machine learning in applications of critical infrastructure such as healthcare and energy, privacy is a growing concern in the minds of stakeholders.

BIG-bench Machine Learning Privacy Preserving +3

Violence Detection in Videos

no code implementations18 Sep 2021 Praveen Tirupattur, Christian Schulze, Andreas Dengel

To address this issue, an approach to automatically detect violent content in videos is proposed in this work.

Binary Classification Classification +2

Spatial Transformer Networks for Curriculum Learning

no code implementations22 Aug 2021 Fatemeh Azimi, Jean-Francois Jacques Nicolas Nies, Sebastian Palacio, Federico Raue, Jörn Hees, Andreas Dengel

Curriculum learning is a bio-inspired training technique that is widely adopted to machine learning for improved optimization and better training of neural networks regarding the convergence rate or obtained accuracy.

Image Classification

A Reinforcement Learning Approach for Sequential Spatial Transformer Networks

no code implementations27 Jun 2021 Fatemeh Azimi, Federico Raue, Joern Hees, Andreas Dengel

Spatial Transformer Networks (STN) can generate geometric transformations which modify input images to improve the classifier's performance.

Decision Making reinforcement-learning +1

AudioCLIP: Extending CLIP to Image, Text and Audio

4 code implementations24 Jun 2021 Andrey Guzhov, Federico Raue, Jörn Hees, Andreas Dengel

AudioCLIP achieves new state-of-the-art results in the Environmental Sound Classification (ESC) task, out-performing other approaches by reaching accuracies of 90. 07% on the UrbanSound8K and 97. 15% on the ESC-50 datasets.

Classification Environmental Sound Classification +2

Understanding Information Spreading Mechanisms During COVID-19 Pandemic by Analyzing the Impact of Tweet Text and User Features for Retweet Prediction

no code implementations26 May 2021 Pervaiz Iqbal Khan, Imran Razzak, Andreas Dengel, Sheraz Ahmed

These social media platforms enable users to share information with other users who can reshare this information, thus causing this information to spread.

Combining Transformer Generators with Convolutional Discriminators

no code implementations21 May 2021 Ricard Durall, Stanislav Frolov, Jörn Hees, Federico Raue, Franz-Josef Pfreundt, Andreas Dengel, Janis Keupe

Transformer models have recently attracted much interest from computer vision researchers and have since been successfully employed for several problems traditionally addressed with convolutional neural networks.

Data Augmentation Image Generation +1

AttrLostGAN: Attribute Controlled Image Synthesis from Reconfigurable Layout and Style

1 code implementation25 Mar 2021 Stanislav Frolov, Avneesh Sharma, Jörn Hees, Tushar Karayil, Federico Raue, Andreas Dengel

In this paper, we propose a method for attribute controlled image synthesis from layout which allows to specify the appearance of individual objects without affecting the rest of the image.

Attribute Layout-to-Image Generation

Deep Learning Based Decision Support for Medicine -- A Case Study on Skin Cancer Diagnosis

no code implementations2 Mar 2021 Adriano Lucieri, Andreas Dengel, Sheraz Ahmed

Moreover, the possibility to intervene and guide models in case of misbehaviour is identified as a major step towards successful deployment of AI as DL-based DSS and beyond.

Confidence-Aware Learning Assistant

no code implementations15 Feb 2021 Shoya Ishimaru, Takanori Maruichi, Andreas Dengel, Koichi Kise

(2) With the help of 20 participants, we observed that correct answer rates of questions were increased by 14% and 17% by giving feedback about correct answers without confidence and incorrect answers with confidence, respectively.

Multiple-choice

Adversarial Text-to-Image Synthesis: A Review

no code implementations25 Jan 2021 Stanislav Frolov, Tobias Hinz, Federico Raue, Jörn Hees, Andreas Dengel

With the advent of generative adversarial networks, synthesizing images from textual descriptions has recently become an active research area.

Adversarial Text Conditional Image Generation +1

Contextual Classification Using Self-Supervised Auxiliary Models for Deep Neural Networks

1 code implementation7 Jan 2021 Sebastian Palacio, Philipp Engler, Jörn Hees, Andreas Dengel

Classification problems solved with deep neural networks (DNNs) typically rely on a closed world paradigm, and optimize over a single objective (e. g., minimization of the cross-entropy loss).

Ranked #89 on Image Classification on CIFAR-100 (using extra training data)

General Classification Image Classification +1

Achievements and Challenges in Explaining Deep Learning based Computer-Aided Diagnosis Systems

no code implementations26 Nov 2020 Adriano Lucieri, Muhammad Naseer Bajwa, Andreas Dengel, Sheraz Ahmed

Remarkable success of modern image-based AI methods and the resulting interest in their applications in critical decision-making processes has led to a surge in efforts to make such intelligent systems transparent and explainable.

Decision Making

The Person Index Challenge: Extraction of Persons from Messy, Short Texts

1 code implementation16 Nov 2020 Markus Schröder, Christian Jilek, Michael Schulze, Andreas Dengel

First, we give a formal definition of the problem and describe a procedure to generate ground truth data for future evaluations.

Leveraging Visual Question Answering to Improve Text-to-Image Synthesis

no code implementations LANTERN (COLING) 2020 Stanislav Frolov, Shailza Jolly, Jörn Hees, Andreas Dengel

We create additional training samples by concatenating question and answer (QA) pairs and employ a standard VQA model to provide the T2I model with an auxiliary learning signal.

Auxiliary Learning Image Generation +2

Hybrid-S2S: Video Object Segmentation with Recurrent Networks and Correspondence Matching

1 code implementation10 Oct 2020 Fatemeh Azimi, Stanislav Frolov, Federico Raue, Joern Hees, Andreas Dengel

In this work, we study an RNN-based architecture and address some of these issues by proposing a hybrid sequence-to-sequence architecture named HS2S, utilizing a dual mask propagation strategy that allows incorporating the information obtained from correspondence matching.

One-shot visual object segmentation Segmentation +3

Benchmarking adversarial attacks and defenses for time-series data

no code implementations30 Aug 2020 Shoaib Ahmed Siddiqui, Andreas Dengel, Sheraz Ahmed

This paves the way for future research in the direction of adversarial attacks and defenses, particularly for time-series data.

Adversarial Defense Benchmarking +2

Training Deep Neural Networks Without Batch Normalization

no code implementations18 Aug 2020 Divya Gaur, Joachim Folz, Andreas Dengel

The main purpose of this work is to determine if it is possible to train networks effectively when batch normalization is removed through adaption of the training process.

Combining Fine- and Coarse-Grained Classifiers for Diabetic Retinopathy Detection

no code implementations28 May 2020 Muhammad Naseer Bajwa, Yoshinobu Taniguchi, Muhammad Imran Malik, Wolfgang Neumeier, Andreas Dengel, Sheraz Ahmed

Visual artefacts of early diabetic retinopathy in retinal fundus images are usually small in size, inconspicuous, and scattered all over retina.

Diabetic Retinopathy Detection

G1020: A Benchmark Retinal Fundus Image Dataset for Computer-Aided Glaucoma Detection

2 code implementations28 May 2020 Muhammad Naseer Bajwa, Gur Amrit Pal Singh, Wolfgang Neumeier, Muhammad Imran Malik, Andreas Dengel, Sheraz Ahmed

Scarcity of large publicly available retinal fundus image datasets for automated glaucoma detection has been the bottleneck for successful application of artificial intelligence towards practical Computer-Aided Diagnosis (CAD).

Optic Cup Segmentation

P2ExNet: Patch-based Prototype Explanation Network

no code implementations5 May 2020 Dominique Mercier, Andreas Dengel, Sheraz Ahmed

Deep learning methods have shown great success in several domains as they process a large amount of data efficiently, capable of solving complex classification, forecast, segmentation, and other tasks.

Time Series Time Series Analysis

Interpreting Deep Models through the Lens of Data

1 code implementation5 May 2020 Dominique Mercier, Shoaib Ahmed Siddiqui, Andreas Dengel, Sheraz Ahmed

Identification of input data points relevant for the classifier (i. e. serve as the support vector) has recently spurred the interest of researchers for both interpretability as well as dataset debugging.

On Interpretability of Deep Learning based Skin Lesion Classifiers using Concept Activation Vectors

no code implementations5 May 2020 Adriano Lucieri, Muhammad Naseer Bajwa, Stephan Alexander Braun, Muhammad Imran Malik, Andreas Dengel, Sheraz Ahmed

This work aims at elucidating a deep learning based medical image classifier by verifying that the model learns and utilizes similar disease-related concepts as described and employed by dermatologists.

Decision Making Image Classification +1

ImpactCite: An XLNet-based method for Citation Impact Analysis

1 code implementation5 May 2020 Dominique Mercier, Syed Tahseen Raza Rizvi, Vikas Rajashekar, Andreas Dengel, Sheraz Ahmed

Therefore, citation impact analysis (which includes sentiment and intent classification) enables us to quantify the quality of the citations which can eventually assist us in the estimation of ranking and impact.

 Ranked #1 on Citation Intent Classification on SciCite (using extra training data)

Citation Intent Classification Classification +5

Explaining AI-based Decision Support Systems using Concept Localization Maps

1 code implementation4 May 2020 Adriano Lucieri, Muhammad Naseer Bajwa, Andreas Dengel, Sheraz Ahmed

We evaluated our proposed method on SCDB as well as a real-world dataset called CelebA.

If You Like It, GAN It. Probabilistic Multivariate Times Series Forecast With GAN

1 code implementation3 May 2020 Alireza Koochali, Andreas Dengel, Sheraz Ahmed

The motivation of the framework is to either transform existing highly accurate point forecast models to their probabilistic counterparts or to train GANs stably by selecting the architecture of GAN's component carefully and efficiently.

Multivariate Time Series Forecasting Probabilistic Time Series Forecasting +1

DeepCFD: Efficient Steady-State Laminar Flow Approximation with Deep Convolutional Neural Networks

2 code implementations19 Apr 2020 Mateus Dias Ribeiro, Abdul Rehman, Sheraz Ahmed, Andreas Dengel

Computational Fluid Dynamics (CFD) simulation by the numerical solution of the Navier-Stokes equations is an essential tool in a wide range of applications from engineering design to climate modeling.

ESResNet: Environmental Sound Classification Based on Visual Domain Models

1 code implementation15 Apr 2020 Andrey Guzhov, Federico Raue, Jörn Hees, Andreas Dengel

Environmental Sound Classification (ESC) is an active research area in the audio domain and has seen a lot of progress in the past years.

Classification Environmental Sound Classification +2

TSInsight: A local-global attribution framework for interpretability in time-series data

no code implementations ICLR 2020 Shoaib Ahmed Siddiqui, Dominique Mercier, Andreas Dengel, Sheraz Ahmed

We approach the problem of interpretability in a novel way by proposing TSInsight where we attach an auto-encoder to the classifier with a sparsity-inducing norm on its output and fine-tune it based on the gradients from the classifier and a reconstruction penalty.

Time Series Time Series Analysis

P $\approx$ NP, at least in Visual Question Answering

1 code implementation26 Mar 2020 Shailza Jolly, Sebastian Palacio, Joachim Folz, Federico Raue, Joern Hees, Andreas Dengel

In recent years, progress in the Visual Question Answering (VQA) field has largely been driven by public challenges and large datasets.

Question Answering Visual Question Answering

Benchmark Performance of Machine And Deep Learning Based Methodologies for Urdu Text Document Classification

no code implementations3 Mar 2020 Muhammad Nabeel Asim, Muhammad Usman Ghani, Muhammad Ali Ibrahim, Sheraz Ahmad, Waqar Mahmood, Andreas Dengel

Second, it investigates the performance impact of traditional machine learning based Urdu text document classification methodologies by embedding 10 filter-based feature selection algorithms which have been widely used for other languages.

Automated Feature Engineering BIG-bench Machine Learning +6

A Hybrid Approach and Unified Framework for Bibliographic Reference Extraction

no code implementations16 Dec 2019 Syed Tahseen Raza Rizvi, Andreas Dengel, Sheraz Ahmed

DeepBiRD was evaluated on two different datasets to demonstrate the generalization of this approach.

SentiCite: An Approach for Publication Sentiment Analysis

no code implementations7 Oct 2019 Dominique Mercier, Akansha Bhardwaj, Andreas Dengel, Sheraz Ahmed

This paper presents a novel system for sentiment analysis of citations in scientific documents (SentiCite) and is also capable of detecting nature of citations by targeting the motivation behind a citation, e. g., reference to a dataset, reading reference.

Sentiment Analysis

A Robust Hybrid Approach for Textual Document Classification

1 code implementation12 Sep 2019 Muhammad Nabeel Asim, Muhammad Usman Ghani Khan, Muhammad Imran Malik, Andreas Dengel, Sheraz Ahmed

Evaluation results reveal that the proposed methodology outperforms the state-of-the-art of both the (traditional) machine learning and deep learning based text document classification methodologies with a significant margin of 7. 7% on 20 Newsgroups and 6. 6% on BBC news datasets.

BIG-bench Machine Learning Classification +5

ProbAct: A Probabilistic Activation Function for Deep Neural Networks

1 code implementation26 May 2019 Kumar Shridhar, Joonho Lee, Hideaki Hayashi, Purvanshi Mehta, Brian Kenji Iwana, Seokjun Kang, Seiichi Uchida, Sheraz Ahmed, Andreas Dengel

We show that ProbAct increases the classification accuracy by +2-3% compared to ReLU or other conventional activation functions on both original datasets and when datasets are reduced to 50% and 25% of the original size.

Image Classification

Interactive Concept Mining on Personal Data -- Bootstrapping Semantic Services

no code implementations14 Mar 2019 Markus Schröder, Christian Jilek, Andreas Dengel

Semantic services (e. g. Semantic Desktops) are still afflicted by a cold start problem: in the beginning, the user's personal information sphere, i. e. files, mails, bookmarks, etc., is not represented by the system.

Management

KINN: Incorporating Expert Knowledge in Neural Networks

no code implementations15 Feb 2019 Muhammad Ali Chattha, Shoaib Ahmed Siddiqui, Muhammad Imran Malik, Ludger van Elst, Andreas Dengel, Sheraz Ahmed

The promise of ANNs to automatically discover and extract useful features/patterns from data without dwelling on domain expertise although seems highly promising but comes at the cost of high reliance on large amount of accurately labeled data, which is often hard to acquire and formulate especially in time-series domains like anomaly detection, natural disaster management, predictive maintenance and healthcare.

Anomaly Detection Management +1

Fusion Strategies for Learning User Embeddings with Neural Networks

no code implementations8 Jan 2019 Philipp Blandfort, Tushar Karayil, Federico Raue, Jörn Hees, Andreas Dengel

In this paper, we run an experiment on movie ratings data, where we analyze the effect on embedding quality caused by several fusion strategies in neural networks.

DeepAnT: A Deep Learning Approach for Unsupervised Anomaly Detection in Time Series

4 code implementations19 Dec 2018 Mohsin Munir, Shoaib Ahmed Siddiqui, Andreas Dengel, Sheraz Ahmed

In contrast to the anomaly detection methods where anomalies are learned, DeepAnT uses unlabeled data to capture and learn the data distribution that is used to forecast the normal behavior of a time series.

Time Series Time Series Anomaly Detection +1

Inflection-Tolerant Ontology-Based Named Entity Recognition for Real-Time Applications

no code implementations5 Dec 2018 Christian Jilek, Markus Schröder, Rudolf Novik, Sven Schwarz, Heiko Maus, Andreas Dengel

Since precision and recall are higher than with comparably fast methods, we conclude that the quality gap between high speed methods and sophisticated NLP pipelines can be narrowed a bit more without losing too much runtime performance.

named-entity-recognition Named Entity Recognition +1

Overcoming Missing and Incomplete Modalities with Generative Adversarial Networks for Building Footprint Segmentation

no code implementations9 Aug 2018 Benjamin Bischke, Patrick Helber, Florian König, Damian Borth, Andreas Dengel

This assumption limits the applications of multi-modal models since in practice the data collection process is likely to generate data with missing, incomplete or corrupted modalities.

Semantic Segmentation

Towards Semantically Enhanced Data Understanding

no code implementations13 Jun 2018 Markus Schröder, Christian Jilek, Jörn Hees, Andreas Dengel

In the field of machine learning, data understanding is the practice of getting initial insights in unknown datasets.

What do Deep Networks Like to See?

1 code implementation CVPR 2018 Sebastian Palacio, Joachim Folz, Jörn Hees, Federico Raue, Damian Borth, Andreas Dengel

To do this, an autoencoder (AE) was fine-tuned on gradients from a pre-trained classifier with fixed parameters.

Image Classification

TSViz: Demystification of Deep Learning Models for Time-Series Analysis

1 code implementation8 Feb 2018 Shoaib Ahmed Siddiqui, Dominik Mercier, Mohsin Munir, Andreas Dengel, Sheraz Ahmed

This is a step towards making informed/explainable decisions in the domain of time-series, powered by deep learning.

Clustering Self-Driving Cars +2

Multi-Task Learning for Segmentation of Building Footprints with Deep Neural Networks

1 code implementation18 Sep 2017 Benjamin Bischke, Patrick Helber, Joachim Folz, Damian Borth, Andreas Dengel

In this paper, we address the problem of preserving semantic segmentation boundaries in high resolution satellite imagery by introducing a new cascaded multi-task loss.

Multi-Task Learning Segmentation +1

Detection of Anomalies in Large Scale Accounting Data using Deep Autoencoder Networks

4 code implementations15 Sep 2017 Marco Schreyer, Timur Sattarov, Damian Borth, Andreas Dengel, Bernd Reimer

Learning to detect fraud in large-scale accounting data is one of the long-standing challenges in financial statement audits or fraud investigations.

Attribute

EuroSAT: A Novel Dataset and Deep Learning Benchmark for Land Use and Land Cover Classification

8 code implementations31 Aug 2017 Patrick Helber, Benjamin Bischke, Andreas Dengel, Damian Borth

We present a novel dataset based on Sentinel-2 satellite images covering 13 spectral bands and consisting out of 10 classes with in total 27, 000 labeled and geo-referenced images.

Earth Observation General Classification +1

Judging a Book By its Cover

4 code implementations28 Oct 2016 Brian Kenji Iwana, Syed Tahseen Raza Rizvi, Sheraz Ahmed, Andreas Dengel, Seiichi Uchida

Book covers communicate information to potential readers, but can that same information be learned by computers?

Genre classification

A Novel Approach for Data-Driven Automatic Site Recommendation and Selection

no code implementations3 Aug 2016 Sebastian Baumbach, Frank Wittich, Florian Sachs, Sheraz Ahmed, Andreas Dengel

The existing approaches for site selection (commonly used by economists) are manual, subjective, and not scalable, especially to Big Data.

A Generic Method for Automatic Ground Truth Generation of Camera-captured Documents

no code implementations4 May 2016 Sheraz Ahmed, Muhammad Imran Malik, Muhammad Zeshan Afzal, Koichi Kise, Masakazu Iwamura, Andreas Dengel, Marcus Liwicki

The method is generic, language independent and can be used for generation of labeled documents datasets (both scanned and cameracaptured) in any cursive and non-cursive language, e. g., English, Russian, Arabic, Urdu, etc.

Optical Character Recognition (OCR)

Symbol Grounding Association in Multimodal Sequences with Missing Elements

no code implementations13 Nov 2015 Federico Raue, Andreas Dengel, Thomas M. Breuel, Marcus Liwicki

We evaluated the proposed extension in the following scenarios: missing elements in one modality (visual or audio) and missing elements in both modalities (visual and sound).

Dynamic Time Warping Missing Elements

Cannot find the paper you are looking for? You can Submit a new open access paper.