Vision-guided object pose estimation for robotic pushing in real-time

Object pushing in robotics has numerous applications, but it often relies on room-bound object tracking systems such as Motion Capture (MoCap) for accurate object pose acquisition. Such systems limit the potential use scenarios, since they add complexity and cost and require expansion of the sensor infrastructure for expanding the operational...

Automatic Detection of Mind-Wandering using Facial Expressions

Spending time in front of screens has become an inescapable activity, which might be interrupted by unrelated external causes. While automatic approaches to identify mind-wandering (MW) have already been investigated, past research was done with self-reports or physiological data. This work explores automated detection utilizing solely facial...

Monitoring the health of urban greenery with terrestrial low-cost, mobile sensors

Urban forests and vegetation are fundamental for developing resilient cities. Thus, the effective management and protection of urban trees and greenery are essential. Nowadays, urban trees are experiencing atypical amount of natural and human-induced stresses which affects their functionality, productivity and survival. The current methods for...

Lane Detection using Spatio-Temporal Attention

Lane detection represents a fundamental task for automated/autonomous vehicles. Current lane detection methods do not provide the versatility of real-time performance, robustness,and accuracy required for real-world scenarios. The reasons include lack of computing power while being portable and inability to observe the continuity and structure...

Using frequency information to improve accuracy of object detectors

This research paper analyses the effect that using frequency information can have on object detectors. The latter are complex networks that learn information about objects from images and are then able to predict the location of these objects in new, unseen images. There are, however, certain datasets that are hard to learn on, partly because...

Zero-shot learning in pick-and-place tasks using neuro-symbolic concept learning

Pick and place systems that operate in a warehouse setting have been studied a lot recently due to the high economic value for e-commerce companies. In this thesis, the focus is on the perception pipeline that performs object recognition given a certain input data stream (typically RGB-D images). Impressive results regarding object recognition...

Weight Swapping: A new method for Supervised Domain Adaptation in Computer Vision using Discrete Optimization

Training Convolutional Neural Network (CNN) models is difficult when there is a lack of labeled training data and no unlabeled data is available. A popular method for this is domain adaptation where the weights of a pre-trained CNN model are transferred to the problem setup. The model is pre-trained on the same task but in a different domain...

Evaluating the performance of the LIME and Grad-CAM explanation methods on a LEGO multi-label image classification task

In this paper, we run two methods of explanation, namely LIME and Grad-CAM, on a convolutional neural network trained to label images with the LEGO bricks that are visible in them. We evaluate them on two criteria, the improvement of the network's core performance and the trust they are able to generate for users of the system. We nd that in...

Addressing Illumination-Based Domain Shifts in Deep Learning: A Physics-Based Approach

This work investigates how prior knowledge from physics-based reflection models can be used to improve the performance of semantic segmentation models under an illumination-based domain shift. We implement various color invariants as a preprocessing step and find that CNNs trained on these color invariants get stuck in worse local minima...

Computer Vision for Gear Alignment Check and Condition Monitoring of Wind Turbine Gearboxes

Wind energy plays a major role in the ongoing energy transition. To accelerate the adoption of wind energy and thereby the energy transition, the Levelized Cost of Energy (LCOE) has to be minimized. Apart from increasing turbine performance, reducing turbine down-time can contribute to lowering the LCOE.Down-time is defined as time during which...

Full-Whisker Tracking System: A new algorithm for accurate whisker tracking in untrimmed head-fixed mice.

The movement of whiskers in head-fixed mice is of high interest for neurological research, as it allows scientists to learn more about learning processes during active touch. However, manual tracking of whiskers in thousands of frames is not feasible, and reliable tracking of individual whiskers is not possible with the best current software...

Deep Reinforcement Learning for Goal-directed Visual Navigation

Safe navigation in a cluttered environment is a key capability for the autonomous operation of Micro Aerial Vehicles (MAVs). This work explores a (deep) Reinforcement Learning (RL) based approach for monocular vision based obstacle avoidance and goal directed navigation for MAVs in cluttered environments. We investigated this problem in the...

High-Speed Object Detection: Design, Study and Implementation of a Detection Framework using Channel Features and Boosting

In this thesis we design, implement and study a high-speed object detection framework. Our baseline detector uses integral channel features as object representation and AdaBoost as supervised learning algorithm. We suggest the implementation of two approximation techniques for speeding up the baseline detector and show their effectiveness by...