Konferenční příspěvky / Conference Papers (KKY) Collection home page

Browse
Subscribe to this collection to receive daily e-mail notification of new additions RSS Feed RSS Feed RSS Feed
Collection's Items (Sorted by Submit Date in Descending order): 1 to 20 of 154
Gruber, Ivan , Krňoul, Zdeněk , Hrúz, Marek , Kanis, Jakub , Boháček, Matyáš
Mutual Support of Data Modalities in the Task of Sign Language Recognition

This paper presents a method for automatic sign language recognition that was utilized in the CVPR 2021 ChaLearn Challenge (RGB track). Our method is composed of several approaches combined in an ensemble scheme to perform isolated sign-gesture recognition. We combine modalities of vide...

Joly, Alexis , Goëau, Hervé , Cole, Elijah , Kahl, Stefan , Picek, Lukáš , Glotin, Hervé , Deneu, Benjamin , Servajean, Maximillien , Lorieul, Titouan , Vellinga, Willem-Pier , Bonnet, Pierre , Durso, Andrew M. , de Castañeda, Rafael Ruiz , Eggel, Ivan , Müller, Henning
LifeCLEF 2021 Teaser: Biodiversity Identification and Prediction Challenges

Building accurate knowledge of the identity, the geographic distribution and the evolution of species is essential for the sustainable development of humanity, as well as for biodiversity conservation. However, the difficulty of identifying plants and animals in the field is hindering the&#x...

Picek, Lukáš , Durso, Andrew M. , Bolon, Isabelle , de Castañeda, Rafael Ruiz
Overview of SnakeCLEF 2021: Automatic snake species identification with country-level focus

A robust and accurate AI-driven system as an assistance tool for snake species identification has vast potential to help lower deaths and disabilities caused by snakebites. With that in mind, we prepared the SnakeCLEF 2021: Automatic Snake Species Identification Challenge with Country-Level&...

Soukup, Lukáš
Automatic Coral Reef Annotation, Localization and Pixel-wise Parsing Using Mask R-CNN

This paper describes the methods that were used for annotation, localization and pixel-wise parsing of the coral reefs from underwater images. The proposed system achieved competitive results in the third edition of ImageCLEFcoral 2021 challenge. Specifically, in case of annotation and local...

Helma, Václav , Goubej, Martin , Šetka, Vlastimil
Inertial measurements processing for sway angle estimation in overhead crane control applications

The main scope of this paper is to propose data fusion algorithms suitable for estimation of gantry crane hook tilt angles based on the MEMS accelerometer and gyroscope readings. Such methods should merge useful information from both these sensors into a better estimate than t...

Helma, Václav , Goubej, Martin
Active anti-sway crane control using partial state feedback from inertial sensor

The paper deals with development of active anti-sway feedback control method for gantry cranes. Inertial measurement unit is chosen as a load motion sensing device allowing to close a feedback loop. The paper provides guidelines for the successive steps of mathematical modelling, data-d...

Joly, Alexis , Goëau, Hervé , Kahl, Stefan , Picek, Lukáš , Lorieul, Titouan , Cole, Elijah , Deneu, Benjamin , Servajean, Maximillien , Durso, Andrew , Bolon, Isabelle , Glotin, Hervé , Planqué, Robert , de Castañeda, Rafael Ruiz , Vellinga, Willem-Pier , Klinck, Holger , Denton, Tom , Eggel, Ivan , Bonnet, Pierre , Müller, Henning
Overview of LifeCLEF 2021: An Evaluation of Machine-Learning Based Species Identification and Species Distribution Prediction

Building accurate knowledge of the identity, the geographic distribution and the evolution of species is essential for the sustainable development of humanity, as well as for biodiversity conservation. However, the difficulty of identifying plants and animals is hindering the aggregation of ...

Chamidullin, Rail , Šulc, Milan , Matas, Jiří , Picek, Lukáš
A deep learning method for visual recognition of snake species

The paper presents a method for image-based snake species identification. The proposed method is based on deep residual neural networks - ResNeSt, ResNeXt and ResNet - fine-tuned from ImageNet pre-trained checkpoints. We achieve performance improvements by: discarding predictions of species that&...

Psutka, Josef , Vaněk, Jan , Pražák, Aleš
Various DNN-HMM architectures used in acoustic modeling with single-speaker and single-channel

In this paper, we discuss some interesting features of training a special acoustic model for only one speaker with a constant acoustic background (acoustic channel). Currently, the LF-MMI method achieves the best results in many speech recognition tasks. A typical LF-MMI training proced...

Vyskočil, Jiří , Picek, Lukáš
Improving web user interface element detection using Faster R-CNN

Several challenges may arise when designing new user interfaces (UIs), e.g., because of communication between designers and developers, to which the detection of UI elements can help. The ImageCLEF DrawnUI 2021 challenge builds on the detection of such elements in two contest tasks:...

Gruber, Ivan , Hrúz, Marek , Železný, Miloš , Karpov, Alexey
X-Bridge: Image-to-Image Translation with Reconstruction Capabilities

This work presents a novel method for image-to-image translation named X-Bridge. The method is based on a conditional adversarial network. X-Bridge is a supervised method build upon the Pix2pix approach, however, it extends the original system with an additional reconstruction path and ...

Ausberger, Tomáš , Kubíček, Karel , Medvecová, Pavla , Myslivec, Tomáš
Test case generation for Function Block Diagram based on blocks’ predefined behaviour

Automatic test case generation based on knowledge of a model is currently a challenge for many researchers and developers. This article describes the first of two complementary methods for test case generation for Function Block Diagram (FBD) models and grey-box testing. The first ...

Matoušek, Jindřich , Tihelka, Daniel
A Comparison of Convolutional Neural Networks for Glottal Closure Instant Detection from Raw Speech

In this paper, we continue to investigate the use of machine learning for the automatic detection of glottal closure instants (GCIs) from raw speech. We compare several deep one-dimensional convolutional neural network architectures on the same data and show that the InceptionV3 model&#...

Kalista, Karel , Liška, Jindřich , Jakl, Jan
A Vibration Sensor-Based Method for Generating the Precise Rotor Orbit Shape with General Notch Filter Method for New Rotor Seal Design Testing and Diagnostics

Verification of the behaviour of new designs of rotor seals is a crucial phase necessary for their use in rotary machines. Therefore, experimental equipment for the verification of properties that have an effect on rotor dynamics is being developed in the test laboratories of ...

Švec, Jan , Šmídl, Luboš , Psutka, Josef , Pražák, Aleš
Spoken Term Detection and Relevance Score Estimation Using Dot-Product of Pronunciation Embeddings

The paper describes a novel approach to Spoken Term Detection (STD) in large spoken archives using deep LSTM networks. The work is based on the previous approach of using Siamese neural networks for STD and naturally extends it to directly localize a spoken term and estim...

Tihelka, Daniel , Řezáčková, Markéta , Grůber, Martin , Hanzlíček, Zdeněk , Vít, Jakub , Matoušek, Jindřich
Save Your Voice: Voice Banking and TTS for Anyone

The paper describes the process of automatic building of a personalized TTS system. The system was primarily developed for people facing the threat of voice loss; however, it can be used by anyone who wants to save his/her voice for any reason. Regarding the target g...

Pražák, Aleš , Loose, Zdeněk , Psutka, Josef , Radová, Vlasta , Psutka, Josef , Švec, Jan
Live TV Subtitling Through Respeaking

In this paper, we describe our solution for live TV subtitling. The subtitling system uses the respeaking concept with respeakers closely tied with the automatic speech recognition system. The ASR is specially tailored to the live subtitling task by using respeaker-specific acoustic mod...

Chýlek, Adam , Švec, Jan , Šmídl, Luboš
Initial Experiments on Question Answering from the Intrinsic Structure of Oral History Archives

Large audio archives with spoken content are natural candidates for question answering systems. Oral history archives generally contain many facts and stories that would be otherwise hard to obtain without listening to hours of recordings. We strive for making the archive more accessibl...

Volín, Jan , Řezáčková, Markéta , Matoušek, Jindřich
Human and Transformer-Based Prosodic Phrasing in Two Speech Genres

The chief objective of the study was to observe phrasing behaviour of transformer-based neural networks from the linguistic point of view. The transformer-based architecture mapped prosodic phrasing in isolated sentences read out on request, but was commanded to predict prosodic phrases in&#...

Bouček, Zdeněk , Neduchal, Petr , Flídr, Miroslav
DronePort: Smart Drone Battery Management System

This paper deals with the description of a drone management system for long-term missions called DronePort. First, the issue of long-term missions and possible approaches are outlined. Further, the individual components of proposed system, both hardware, and software are introduced. The Dron...

Collection's Items (Sorted by Submit Date in Descending order): 1 to 20 of 154