Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 4 of 4 for “"CNN+RNN"”.
-
Development of an Emotion Recognition Classifier from Body Language Using Deep Learning for the Children with Autism to Help Identifying Human Emotions
… from body language: Convolution Neural Network (CNN), the combination of CNN and Recurrent Neural Network (CNN+RNN), and the combination of CNN and Long Short-Term Memory (CNN+LSTM). For this research, though the CNN model provides better accuracy than CNN+RNN and CNN+LSTM, CNN model ignores the …
-
Deep neural networks for video classification in ecology
… is to compare the performance of image-only CNNs, which treat each frame of a video independently, against a combined CNN-RNN approach; and to assess whether incorporating the motion information in the temporal aspect of video improves the accuracy of classifications in these two datasets. …
-
Prořezávání hlubokých neuronových sítí pro rozpoznávání textu
Dokument predstavuje prácu na prerezávanie neurónovej siete slúžiacej na rozpoznávanie ručne písaného textu. Cieľom práce je vytvoriť program na prerezávanie danej siete. Prerezávali sme dva typy neurónových sieti a to konvolučné a rekurentné neurónové siete. Pri prerezávaní konvolučnej časti bolo …
-
Applying Neural Networks for Tire Pressure Monitoring Systems
… network with long short-term memory blocks (RNN-LSTM) and a convolutional neural network (CNN) developed in Python with Tensorflow. Bayesian Optimization via SigOpt was used to optimize training and model parameters. The predictive accuracy and training speed of the two models with various …