Pieter Simoens
YOU?
Author Swipe
View article: Privacy-Preserving Computer Vision for Industry: Three Case Studies in Human-Centric Manufacturing
Privacy-Preserving Computer Vision for Industry: Three Case Studies in Human-Centric Manufacturing Open
The adoption of AI-powered computer vision in industry is often constrained by the need to balance operational utility with worker privacy. Building on our previously proposed privacy-preserving framework, this paper presents its first com…
View article: Modulation of temporal decision-making in a deep reinforcement learning agent under the dual-task paradigm
Modulation of temporal decision-making in a deep reinforcement learning agent under the dual-task paradigm Open
This study explores the interference in temporal processing within a dual-task paradigm from an artificial intelligence (AI) perspective. In this context, the dual-task setup is implemented as a simplified version of the Overcooked environ…
View article: In-Field Mapping of Grape Yield and Quality With Illumination-Invariant Deep Learning
In-Field Mapping of Grape Yield and Quality With Illumination-Invariant Deep Learning Open
This paper presents an end-to-end, IoT-enabled robotic system for the non-destructive, real-time, and spatially-resolved mapping of grape yield and quality (Brix, Acidity) in vineyards. The system features a comprehensive analytical pipeli…
View article: Source-Free Model Transferability Assessment for Smart Surveillance via Randomly Initialized Networks
Source-Free Model Transferability Assessment for Smart Surveillance via Randomly Initialized Networks Open
Smart surveillance cameras are increasingly employed for automated tasks such as event and anomaly detection within smart city infrastructures. However, the heterogeneity of deployment environments, ranging from densely populated urban int…
View article: Humans program artificial delegates to accurately solve collective-risk dilemmas but lack precision
Humans program artificial delegates to accurately solve collective-risk dilemmas but lack precision Open
In an era increasingly influenced by autonomous machines, it is only a matter of time before strategic individual decisions that impact collective goods will also be made virtually through the use of artificial delegates. Through a series …
View article: Adaptive Clustering for Efficient Phenotype Segmentation of UAV Hyperspectral Data
Adaptive Clustering for Efficient Phenotype Segmentation of UAV Hyperspectral Data Open
Unmanned Aerial Vehicles (UAVs) combined with Hyperspectral imaging (HSI) offer potential for environmental and agricultural applications by capturing detailed spectral information that enables the prediction of invisible features like bio…
View article: Embedding-based pair generation for contrastive representation learning in audio-visual surveillance data
Embedding-based pair generation for contrastive representation learning in audio-visual surveillance data Open
Smart cities deploy various sensors such as microphones and RGB cameras to collect data to improve the safety and comfort of the citizens. As data annotation is expensive, self-supervised methods such as contrastive learning are used to le…
View article: Enabling Privacy-Aware AI-Based Ergonomic Analysis
Enabling Privacy-Aware AI-Based Ergonomic Analysis Open
Musculoskeletal disorders (MSDs) are a leading cause of injury and productivity loss in the manufacturing industry, incurring substantial economic costs. Ergonomic assessments can mitigate these risks by identifying workplace adjustments t…
View article: Hybrid Edge–Cloud Models for Bearing Failure Detection in a Fleet of Machines
Hybrid Edge–Cloud Models for Bearing Failure Detection in a Fleet of Machines Open
Real-time condition monitoring of machinery is increasingly being adopted to minimize costs and enhance operational efficiency. By leveraging large-scale data acquisition and intelligent algorithms, failures can be detected and predicted, …
View article: Predicting change in time production -- A machine learning approach to time perception
Predicting change in time production -- A machine learning approach to time perception Open
Time perception research has advanced significantly over the years. However, some areas remain largely unexplored. This study addresses two such under-explored areas in timing research: (1) A quantitative analysis of time perception at an …
View article: Reward Machine Inference for Robotic Manipulation
Reward Machine Inference for Robotic Manipulation Open
Learning from Demonstrations (LfD) and Reinforcement Learning (RL) have\nenabled robot agents to accomplish complex tasks. Reward Machines (RMs) enhance\nRL's capability to train policies over extended time horizons by structuring\nhigh-le…
View article: Learning Task Specifications from Demonstrations as Probabilistic Automata
Learning Task Specifications from Demonstrations as Probabilistic Automata Open
Specifying tasks for robotic systems traditionally requires coding expertise, deep domain knowledge, and significant time investment. While learning from demonstration offers a promising alternative, existing methods often struggle with ta…
View article: Revisiting Edge AI: Opportunities and Challenges
Revisiting Edge AI: Opportunities and Challenges Open
Edge artificial intelligence (AI) is an innovative computing paradigm that aims to shift the training and inference of machine learning models to the edge of the network. This paradigm offers the opportunity to significantly impact our eve…
View article: Reactive shepherding along a dynamic path
Reactive shepherding along a dynamic path Open
View article: Multi-camera detection framework for lifelong broiler flock monitoring
Multi-camera detection framework for lifelong broiler flock monitoring Open
In recent times, there has been an increased emphasis on chicken welfare within the poultry industry, leading to the adoption of camera-based solutions for continuous monitoring of broilers' activity and health parameters. Accurate detecti…
View article: Mitigating Bias Using Model-Agnostic Data Attribution
Mitigating Bias Using Model-Agnostic Data Attribution Open
Mitigating bias in machine learning models is a critical endeavor for ensuring fairness and equity. In this paper, we propose a novel approach to address bias by leveraging pixel image attributions to identify and regularize regions of ima…
View article: Test-time Specialization of Dynamic Neural Networks
Test-time Specialization of Dynamic Neural Networks Open
In recent years, there has been a notable increase in the size of commonly used image classification models. This growth has empowered models to recognize thousands of diverse object types. However, their computational demands pose signifi…
View article: Label Efficient Lifelong Multi-View Broiler Detection
Label Efficient Lifelong Multi-View Broiler Detection Open
Broiler localization is crucial for welfare monitoring, particularly in identifying issues such as wet litter. We focus on multi-camera detection systems since multiple viewpoints not only ensure comprehensive pen coverage but also reduce …
View article: Multi-bit, Black-box Watermarking of Deep Neural Networks in Embedded Applications
Multi-bit, Black-box Watermarking of Deep Neural Networks in Embedded Applications Open
The effort required to collect data and train a large neural network requires a significant investment from organizations. Therefore, trained neural networks are often seen as valuable intellectual property that needs to be protected. At t…
View article: Mitigating Bias Using Model-Agnostic Data Attribution
Mitigating Bias Using Model-Agnostic Data Attribution Open
Mitigating bias in machine learning models is a critical endeavor for ensuring fairness and equity. In this paper, we propose a novel approach to address bias by leveraging pixel image attributions to identify and regularize regions of ima…
View article: Committing to the wrong artificial delegate in a collective-risk dilemma is better than directly committing mistakes
Committing to the wrong artificial delegate in a collective-risk dilemma is better than directly committing mistakes Open
View article: Privacy-preserving visual analysis: training video obfuscation models without sensitive labels
Privacy-preserving visual analysis: training video obfuscation models without sensitive labels Open
View article: The art of compensation: How hybrid teams solve collective-risk dilemmas
The art of compensation: How hybrid teams solve collective-risk dilemmas Open
It is widely known how the human ability to cooperate has influenced the thriving of our species. However, as we move towards a hybrid human-machine future, it is still unclear how the introduction of artificial agents in our social intera…
View article: Multi-Camera Detection Framework for Lifelong Broiler Flock Monitoring
Multi-Camera Detection Framework for Lifelong Broiler Flock Monitoring Open
View article: Learning Safety Constraints From Demonstration Using One-Class Decision Trees
Learning Safety Constraints From Demonstration Using One-Class Decision Trees Open
The alignment of autonomous agents with human values is a pivotal challenge when deploying these agents within physical environments, where safety is an important concern. However, defining the agent's objective as a reward and/or cost fun…
View article: The duplication of genomes and genetic networks and its potential for evolutionary adaptation and survival during environmental turmoil
The duplication of genomes and genetic networks and its potential for evolutionary adaptation and survival during environmental turmoil Open
The importance of whole-genome duplication (WGD) for evolution is controversial. Whereas some view WGD mainly as detrimental and an evolutionary dead end, there is growing evidence that polyploidization can help overcome environmental chan…
View article: Cyclic Action Graphs for goal recognition problems with inaccurately initialised fluents
Cyclic Action Graphs for goal recognition problems with inaccurately initialised fluents Open
Goal recognisers attempt to infer an agent’s intentions from a sequence of observed actions. This is an important component of intelligent systems that aim to assist or thwart actors; however, there are many challenges to overcome. For exa…
View article: The Effect of Rapport on Delegation to Virtual Agents
The Effect of Rapport on Delegation to Virtual Agents Open
This paper presents the initial results of a study exploring whether the perceived rapport with a virtual agent can influence users' decisions on delegating critical tasks to the agent. We hypothesize that users are more likely to delegate…
View article: Maximum Causal Entropy Inverse Constrained Reinforcement Learning
Maximum Causal Entropy Inverse Constrained Reinforcement Learning Open
When deploying artificial agents in real-world environments where they interact with humans, it is crucial that their behavior is aligned with the values, social norms or other requirements of that environment. However, many environments h…
View article: Steering herds away from dangers in dynamic environments
Steering herds away from dangers in dynamic environments Open
Shepherding, the task of guiding a herd of autonomous individuals in a desired direction, is an essential skill to herd animals, enable crowd control and rescue from danger. Equipping robots with the capability of shepherding would allow p…