Exploring Computer Vision: From

Fundamentals to Advanced
Applications
Katende Jericho 21/U/1565
Musemeza Murungi Isaac 21/U/1395
Mawejje Mark William 21/U/09946/EVE
Sserubidde Aisha 21/U/15126/PS
Barasa Elijah Maiga 21/U/06647/PS
JESUS I TRUST IN YOU
001

002
003
Introduction to Computer Vision
004
005
006
Objective: Introduce the concept of computer vision, its applications, and
007 provide a timeline for the subsequent presentations in the series.

008
009
010
011
001
002

003
Definition:
004
Computer vision is a field of artificial intelligence
005 (AI) that enables computers and systems to
derive meaningful information from digital
006 images, videos and other visual inputs — and
take actions or make recommendations based on
007 that information. If AI enables computers to
think, computer vision enables them to see,
008 observe and understand.

009
010
011
001
002
003
How Computer Vision Works
004 Computer vision needs lots of data. It runs analyses of data over and over until it discerns
distinctions and ultimately recognize images.
005
It accomplishes this by use of Convolutional Neural Network(CNN): CNNs are a specific
006 type of deep neural network designed for processing structured grid data, such as images
and videos.
007
008
009
010
011
001
How Computer Vision Works
002
003 Step Description

Image Acquisition Acquiring digital images or videos using cameras, sensors, or other imaging
004 devices.
Preprocessing Enhancing image quality, removing noise, adjusting lighting conditions, or resizing
005 images.
Feature Extraction Extracting meaningful features like edges, shapes, textures, colors, or other visual
006 patterns.
Feature Representation Converting extracted features into numerical values or feature vectors.
007
Pattern Recognition Analyzing extracted features to identify patterns or objects using machine learning
008 models.
Decision Making Making decisions or taking actions based on identified patterns or objects in the
009 images.
Feedback Loop Incorporating feedback to improve system performance through retraining or

010 parameter adjustment.

011
001
002
003 Historical Perspective
1960s: Initial research and development of algorithms to process and analyze visual data begins.
004 1970s: More sophisticated algorithms for image processing and pattern recognition emerge. Hough transform
facilitates line detection.
005
1980s-1990s: Focus shifts to developing machine learning algorithms for computer vision. Viola-Jones face
detection algorithm is a major breakthrough.
006
2000s-2010: Deep learning revolutionizes computer vision. CNNs and other deep learning algorithms enable

007 hierarchical representations of visual data.

Present and Future:

008 Rapid Evolution: Computer vision is rapidly advancing due to breakthroughs in artificial intelligence and deep
learning.
009 Integration with Emerging Tech: Integration with technologies like AR, VR, and IoT expands its applications
across various industries.
Ethical Concerns: As computer vision proliferates, ethical considerations regarding bias and privacy become
010 paramount.

011
001
002
003 Notable Researchers in this Field
1. Andrew Ng: Co-founder of deeplearning.ai and Coursera, known for contributions to
004 machine learning and computer vision.Leads of google brain.
2. Fei-Fei Li: Sequoia Professor at Stanford University, key figure in computer vision and
005 AI diversity advocacy.

006 3. Andrej Karpathy: Senior Director of AI at Tesla, expertise in deep learning and
computer vision.

007 4. Demis Hassabis: Co-founder and CEO of DeepMind, leader in AI research and
applications.

008 5. Ian Goodfellow: Director of Machine Learning at Apple, known for GANs and deep
learning research.
009 6. Yann LeCun: Chief AI Scientist at Facebook, pioneering work in convolutional nets and
AI research.
010 7. Jeremy Howard: Founding Researcher at fast.ai, leading figure in AI education and
entrepreneurship.
011
001
002 Applications of Computer Vision

003 Application Description

004 Autonomous Vehicles Enables self-driving cars to perceive and interpret surroundings, detect obstacles,
recognize traffic signs, and navigate.

005 Healthcare Utilized for medical image analysis, tumor detection, pathology diagnosis, and
surgery assistance.

006 Retail Applied in inventory management, shelf stocking optimization, customer behavior
analysis, and facial recognition for payments. (Amazon Go).

007 Surveillance and Security Used for video surveillance, real-time monitoring, object detection, facial
recognition, and anomaly detection.

008 Augmented Reality (AR) and Virtual

Reality (VR)
Essential for depth perception, object recognition, and interaction with virtual
elements in real-time.

009 Manufacturing Supports quality control through automated inspection, defect detection, and
ensuring product consistency.

010 Agriculture Assists in crop monitoring, yield prediction, and disease detection in plants,
optimizing agricultural processes.

011
001
CONTINUATION
002
Education Used for student engagement analysis, monitoring classroom behavior, and automating grading processes.
003 Gesture Recognition Interprets human gestures for touchless interactions in devices and control systems.

004 Retail Analytics Applied for customer tracking, heatmaps, and demographic analysis in retail spaces to optimize store
layouts.

005 Document Analysis Optical Character Recognition (OCR) systems extract text and information from images or scanned
documents.

006 Entertainment Contributes to facial recognition in gaming, emotion analysis, and character animation for an enhanced
entertainment experience.

007 Environmental Monitoring Used for monitoring environmental conditions, wildlife tracking, and analyzing satellite imagery for
climate studies.

008 Sports Analytics Employed for player tracking, performance analysis, and generating statistics to improve training and
game strategies.

009 Smart Cities Contributes to traffic management, public safety, waste management, and urban planning in the context of
developing smart cities.

010
011
001
002
003 Closing Thoughts on the Future of Computer Vision; Continuing the
Exploration.
004
005 Continued research will enhance computer vision, enabling broader image detection and easier
training. Integration with Natural language Generation (NLG) and other AI fields will create
006 versatile applications, aiding visually impaired individuals. Overall, computer vision's
evolution promises increased functionality and agility in diverse contexts.

007
008
009

010
011
001
002
003 Course Roadmap
004 ● Outline the topics to be covered in subsequent presentations:
● Presentation 2: Image Processing Techniques
005 ● Presentation 3: Feature Extraction and Object Detection
● Presentation 4: Deep Learning for Computer Vision

006 ● Presentation 5: Advanced Topics in Computer Vision

● Presentation 6: Future Trends and Challenges

007
008
009
010

011
001
002
003 References

004 https://www.sama.com/blog/computer-vision-histo
ry-how-it-works/#:~:text=The%20History%20of%20
005 Computer%20Vision,image%20processing%20and%
20pattern%20recognition
006 .
007 https://www.mygreatlearning.com/blog/ai-research
ers-and-leaders/
008 https://addepto.com/blog/the-future-of-computer-
009 vision-and-artificial-intelligence/#
https://www.mygreatlearning.com/blog/what-is-
010
computer-vision-the-basics/#:~:text=Computer
011 %20Vision%20primarily%20relies%20on,process
%20more%20accurate%20and%20fast.