Scribd
Upload
7 views

CV 01 Introduction

The document provides an introduction to computer vision including defining it, discussing its history and applications, challenges, and instructions for a basic implementation. Computer vision is a field of artificial intelligence that enables computers to derive meaning from images and videos. It has various applications including facial recognition, medical imaging, and autonomous vehicles. Key challenges include variations in lighting, scale, occlusion and complexity of real-world scenes.

Uploaded by

Nguyễn Bin
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
7 views

CV 01 Introduction

The document provides an introduction to computer vision including defining it, discussing its history and applications, challenges, and instructions for a basic implementation. Computer vision is a field of artificial intelligence that enables computers to derive meaning from images and videos. It has various applications including facial recognition, medical imaging, and autonomous vehicles. Key challenges include variations in lighting, scale, occlusion and complexity of real-world scenes.

Uploaded by

Nguyễn Bin
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

COMPUTER VISION

INTRODUCTION

Ngo Quoc Viet-2024


Contents

NGO QUOC VIET


• Introduction

• Brief history of computer vision

• Some applications of computer vision

• The challenges in computer vision.

• Getting started with tiny installation/implementation

2
Lecture outcomes

NGO QUOC VIET


• Understanding what are computer vision, some its applications and main
challenges.

• Implementing the getting started tiny application in CV by loading, showing


image/video and putting your brand (text or icon) on the frames.

3
What is computer vision

NGO QUOC VIET


• Computer vision is a field of artificial intelligence (AI) that enables computers and systems to derive meaningful
information from digital images, videos and other visual inputs-and take actions or make recommendations
based on that information(https://www.ibm.com/topics/computer-vision).

• Computer vision is a field of computer science that focuses on enabling computers to identify and understand
objects and people in images and videos. Like other types of AI, computer vision seeks to perform and automate
tasks that replicate human capabilities. In this case, computer vision seeks to replicate both the way humans
see, and the way humans make sense of what they see (https://azure.microsoft.com/en-us/resources/cloud-
computing-dictionary/what-is-computer-vision#object-classification).

• Computer vision, a fascinating field at the intersection of computer science and artificial intelligence, which
enables computers to analyze images or video data, unlocking a multitude of applications across industries, from
autonomous vehicles to facial recognition systems (https://www.geeksforgeeks.org/computer-vision/).
4
Computer vision and related disciplines

NGO QUOC VIET


https://www.researchgate.net/figure/Relationship-between-computer-vision-and-related-disciplines_fig2_360146441 5
Brief history of computer vision

NGO QUOC VIET


https://youtu.be/qHvsocj1c08 6
Brief history of computer vision

NGO QUOC VIET


• 1960s: Researchers aimed to enable computers to interpret visual information. MIT AI
lab begins computer vision projects.

• 1970s: Optical Character Recognition (OCR) debuts via Kurzweil Computer Products.

• 1980s: Smart cameras, such as the Xerox optical mouse, and layered image
processing.

• 2000s: Viola-Jones object detection framework (facial recognition and object


detection); Pascal VOC Project launched for developing image datasets for training.

7
Brief history of computer vision

NGO QUOC VIET


• 2005: Security Show Japan 2005 and the first mobile phone with working facial
recognition features.

• 2010: AI’s average object-recognition error rate becomes better than humans’; 2.5%
versus 5%; First ImageNet competition. Developers competed for developing image
recognition models. ImageNet now has up to 15 million images in 20,000 categories.

• 2015: Google introduces FaceNet for facial recognition. Requires minimum data set.

• 2018: NVIDIA announces that it will make its hyper-realistic face generator StyleGAN
open source.
8
Some applications of computer vision

NGO QUOC VIET


9
Some applications of computer vision

NGO QUOC VIET


10
Some applications of computer vision

NGO QUOC VIET


11
The challenges in computer vision

NGO QUOC VIET


• Variations in Lighting: Changes in lighting conditions can
drastically alter the appearance of an object, making it
difficult for computer vision systems to consistently
recognize objects.
• Changes in Perspective: While humans are generally
good at recognizing objects from different perspectives,
this is a significant challenge for computer vision systems. Some of the lighting variations that occurred in
one day of testing. Changes in lighting added
• Scale Invariance: Another challenge is recognizing complexity for computer vision due to shadows
objects of different sizes or when the size of the object and reflections on the food tables

changes due to changes in distance

12
The challenges in computer vision

NGO QUOC VIET


• Occlusion: Occlusion occurs when an object of interest is
partially hidden by other objects. Recognizing an object
when only part of it is visible is a complex problem
• Scene Complexity: Real-world scenes can be incredibly
complex, with many objects interacting in various ways.
Understanding these scenes, especially when the objects
are similar or overlapping.
handling complex scenes with
• Real-Time Processing: Many applications, such as
occlusions and cluttered backgrounds,
autonomous driving or video surveillance, require real- scale and perspective variations, object
time processing class imbalance, and achieving real-
time performance on resource-
constrained devices

13
Getting started with tiny application

NGO QUOC VIET


• Capture video or read MP4 from device

• Override your name/icon at the left/right top corner of the frames

• Convert the frames to gray space

14

You might also like