Book Image

Artificial Vision and Language Processing for Robotics

By : Álvaro Morena Alberola, Gonzalo Molina Gallego, Unai Garay Maestre
Book Image

Artificial Vision and Language Processing for Robotics

By: Álvaro Morena Alberola, Gonzalo Molina Gallego, Unai Garay Maestre

Overview of this book

Artificial Vision and Language Processing for Robotics begins by discussing the theory behind robots. You'll compare different methods used to work with robots and explore computer vision, its algorithms, and limits. You'll then learn how to control the robot with natural language processing commands. You'll study Word2Vec and GloVe embedding techniques, non-numeric data, recurrent neural network (RNNs), and their advanced models. You'll create a simple Word2Vec model with Keras, as well as build a convolutional neural network (CNN) and improve it with data augmentation and transfer learning. You'll study the ROS and build a conversational agent to manage your robot. You'll also integrate your agent with the ROS and convert an image to text and text to speech. You'll learn to build an object recognition system using a video. By the end of this book, you'll have the skills you need to build a functional application that can integrate with a ROS to extract useful information about your environment.
Table of Contents (12 chapters)
Artificial Vision and Language Processing for Robotics
Preface

Summary


Conversational agents, also knowns as chatbots, are text-based dialogue systems that understand human language in order to hold a "real" conversation with people. To achieve a good understanding of what a human is saying, chatbots need to classify dialogue into intents, that is, a set of sentences representing a meaning. Conversational agents can be classified into several groups, depending on the type of input-output data and knowledge limits. This representation of meaning is not easy. To have sound knowledge supporting a chatbot, a huge corpus is needed. Finding the best way to represent a word is a challenge, and one-hot encoding is useless. The main problem with one-hot encoding is the size of the encoded vectors. If we have a corpus of 88,000 words, then the vectors will have a size of 88,000, and without any relationship between the words. This is where the concept of word embeddings enters the picture.

Word embeddings are a collection of techniques and methods to map words...