[Pycon] [new paper] "Daniel Ajisafe" - Introduction to deep learning for computer vision

[info a pycon.it]

Title: Introduction to deep learning for computer vision
Duration: 45 (includes Q&A)
Q&A Session: 0
Language: en
Type: Talk

Abstract: The amount of visual data in the world today has grown exponentially in the last couple of years and this is largely due to lots of sensors everywhere. Building machine learning models that are capable of localizing and identifying objects in a single image remains a core challenge in computer vision.

With the rise of convolutional neural networks, increasing amount of data and fast computing power, different deep learning algorithms are being used to solve conventional artificial intelligence problems in image classification, object detection, image extraction and semantic segmentation.

In the first half of this talk, some of the core mathematical concepts and architectures will be explained. The audience will understand why this concepts are essential and why we use them. The second half will include a code demo to show how easy it is to get started using a real world example. From this, the attendees would learn how keras, python and Tensorflow can be used to build convolutional neural networks.

At the end of this talk, the audience would understand the challenges for constructing and training deep neural networks and the potentials they have for the future. 

Tags: [u'neural network', u'computer-vision']