Deep Learning Applications for Computer Vision – Coursera

This course is available on Coursera from the University of Colorado Boulder. Computer vision as a topic of study and research is the emphasis of the curriculum. First, the students will study how computer vision tasks are carried out and the methodologies for model development that are recommended. Following that, the students will learn about deep learning techniques and how to apply them to computer vision challenges. The students will also learn to examine the results as well as the benefits and drawbacks of various strategies.

Finally, the students will participate in hands-on training to learn how to develop models using machine learning tools and libraries. Using advanced neural networks constructed from the ground up, the students will conduct image classification, object identification, object segmentation, facial recognition, and activity and position estimation tasks.

The course modules are:

Introduction and Background

• The first module introduces the field of computer vision and how information can be extracted from images. Furthermore, the learners will cover the primary categories of tasks in computer vision. In addition, the learners will understand how deep learning techniques impact the field of computer vision.

Classic Computer Vision Tools

• The second module allows learners to explore various computer vision tools and techniques, and concepts on convolution operation, linear filters, and algorithms for feature detection and image detection.

Image Classification in Computer Vision

• In the third module, the learners will review the challenges in object recognition in the classic computer vision approach. Next, the learners will understand the necessary steps to perform object recognition and image classification tasks using the computer vision pipeline.

Neural Networks and Deep Learning

• The fourth module focuses on the image classification pipeline using neural networks. The learners will understand the differences between common problems and computer vision problems with neural network models.
• The essential components of neural networks are covered in-depth and practical sessions cover the implementation of neural networks for image classification using TensorFlow.

CNN and Deep Learning Advanced Tools

• The final module provides learners with the concepts of various CNN components, including parameters and hyperparameters in a deep neural network. The learners will also cover essential concepts on improving the accuracy of the deep learning models and the essential factors in improving the overall performance of these models. Furthermore, the project in this module requires learners to build and train a deep neural network for image classification.

Instructor: Ioana Fleming

Level: Intermediate

Duration: 22 hours

User Review: NA

No. of Reviews: NA

Price: Free Enrollment (Additional charges for certification may apply)

Website: coursera.org/learn/deep-learning-computer-vision