Combining Local and Global Pose Estimation for Precise Tracking of Similar Objects
We present a multi-object 6D detection and tracking pipeline for potentially similar and non-textured objects. The combination of a convolutional neural network for object classification and rough pose estimation with a local pose refinement and...
Overview of the Neural Network Compression and Representation (NNR) Standard
Neural Network Coding and Representation (NNR) is the first international standard for efficient compression of neural networks. The NNR standard contains quantization and an arithmetic coding scheme as core encoding and decoding technologies, as...
Automated Damage Inspection of Power Transmission Towers from UAV Images
This paper adresses the problem of structural damage detection in transmission towers, addressing these two common challenges: (i) the lack of freely available training data and the difficulty to collect it; (ii) fuzzy boundaries of what...
Continuous wave terahertz receivers with 4.5 THz bandwidth and 112 dB dynamic range
We present photomixers made of InGaAs:Fe as broadband receivers in optoelectronic cw THz systems. The improved resistivity and carrier lifetime of InGaAs:Fe enable us to measure a bandwidth of 4.5 THz with a peak dynamic range of 112 dB. When...
ML-assisted QoT estimation: a dataset collection and data visualization for dataset quality evaluation
We present a publicly available dataset collection to open the problem of data-driven QoT estimation to the ML community. The dataset collection allows comparing various solutions presented by different research groups. Furthermore, we propose...
Explaining Machine Learning Models for Clinical Gait Analysis
This article investigates the usefulness of Explainable Artificial Intelligence (XAI) methods to increase transparency in automated clinical gait classification based on time series. For this purpose, predictions of state-of-the-art...
Finding and Removing Clever Hans: Using Explanation Methods to Debug and Improve Deep Models
Contemporary learning models for computer vision are typically trained on very large (benchmark) datasets with millions of samples. These may, however, contain biases, artifacts, or errors that have gone unnoticed and are exploitable by the...
Terahertz Multilayer Thickness Measurements: Comparison of Optoelectronic Time and Frequency Domain Systems
We compare a state-of-the-art terahertz (THz) time domain spectroscopy (TDS) system and a novel optoelectronic frequency domain spectroscopy (FDS) system with respect to their performance in layer thickness measurements on dielectric samples....
Inverse kinematics for full-body self representation in VR-based cognitive rehabilitation
Being self-represented through an avatar increases embodiment and the feeling of presence in virtual reality. Nevertheless, currently users in VR are typically represented only by their hands, as not enough tracking data is available for full...
Accurate human body reconstruction for volumetric video
In this work, we enhance a professional end-to-end volumetric video production pipeline to achieve high-fidelity human body reconstruction using only passive cameras.We introduce and optimize deep learning based multi-view stereo networks for...