Browsing 2020 International Conference "Electrical, Electronic and Information Sciences“ (eStream)  by Department "Informacinių technologijų katedra / Department of Information Technologies"

Now showing items 1-3 of 3

    • Comparison of VGG and ResNet used as Encoders for Image Captioning 

      Atliha, Viktar; Šešok, Dmitrij (2020 IEEE Open Conference of Electrical, Electronic and Information Sciences (eStream), April 30, 2020, Vilnius, Lithuania, 2020)
      Recent models for image captioning are usually based on an encoder-decoder framework. Large pre-trained convolutional neural networks are often used as encoders. However, different authors use different encoder architectures ...

    • Investigation of Acoustic Features for Voice Activation Problem 

      Kolesau, Aliaksei; Šešok, Dmitrij (2020 IEEE Open Conference of Electrical, Electronic and Information Sciences (eStream), April 30, 2020, Vilnius, Lithuania, 2020)
      In this paper we examine the results of using different acoustic feature computation pipelines for classifying audio keywords with a convolutional neural network (CNN). We compare the use of Mel-frequency cepstral coefficients ...

    • Near-optimal Placement of Piles in Grillage Type Foundations Based on Simulated Annealing 

      Miloš, Ervin; Kovalyov, Mikhail Y.; Mačiūnas, Darius; Belevičius, Rimantas; Šešok, Dmitrij (2020 IEEE Open Conference of Electrical, Electronic and Information Sciences (eStream), April 30, 2020, Vilnius, Lithuania, 2020)
      A problem of optimal positioning of piles in grillage type foundations is studied. A Global Optimization model reported in the literature is used to formalize this problem. A novel Simulated Annealing algorithm is developed ...