1- Rapid advances in computer vision is enabling brand new applications the were not possible a few years ago.

2- Since the community is creative, it can inspire applications in other areas.

Why computer vision is interesting?

What is the problem of using deep learning on images?

1- Image classification

2- Image localization

3- Object detection

4- Neural style transfer

Types of Computer Vision Problems

Acquired data is usually messy and can come from different sources, angles, color scales, etc. In order to use machine and deep learning techniques they need to be standardized. Usually the goal is to increase the accuracy while reducing the complexity of the model used. There is no single answer to preprocessing but there are a group of techniques commonly chosen from.

Computer Vision Pre-Processing

Computer vision uses many established datasets to train and benchmark algorithms for tasks such as image classification, object detection, segmentation, and pose estimation. Because gathering and annotating large image collections is labor-intensive, the field tends to reuse a relatively small set of standardized benchmark datasets—for example ImageNet, CIFAR-10, and MS COCO—across many studies.

Computer Vision Datasets

- Human computer interaction (HCI)
- Interaction design
- Collaborative and social computing
- Ubiquitous and mobile computing
- Data Visualization
- Accessibility

Human-centered computing in ACM Computing Classification

As computer vision methods have become relatively standardized, it is efficient to utilize established software libraries and tools. Prominent examples include OpenCV, the SciPy ndimage Module, TensorFlow, PyTorch Vision, and the MATLAB Vision Toolbox.

Computer Vision Libraries

University of Michigan - Ann Arbor

- Symbolic and algebraic manipulation
- Parallel computing methodologies
- Artificial intelligence
- Machine learning
- Modeling and simulation
- Computer graphics
- Distributed computing methodologies
- Concurrent computing methodologies

Computing methodologies in ACM Computing Classification

A helpful website for understanding how CNN works:
https://towardsdatascience.com/a-comprehensive-guide-to-convolutional-neural-networks-the-eli5-way-3bd2b1164a53

CNN Reference

AI is the field of study that incorporates human intelligence onto computer systems in the pursuit of making them capable of performing tasks that require human intelligence.

Artifical Intelligence

In the dominant machine learning approach, rather than designing explicit step-by-step rules to map inputs to outputs, practitioners define a flexible program whose behavior is governed by a set of adjustable parameters. By presenting this program with a large dataset of examples and labels, a learning algorithm determines the optimal parameter values that maximize the model's performance on the desired task.

Learn Before

Related

Learn After