Embedded Files
DERIVED GRADE EXAM
DERIVED GRADE EXAM
13COMP
13COMP
Date: tba
Date: tba
Time: tba
Time: tba
DURATION: 2hrs to complete
DURATION: 2hrs to complete
ROOM: tba
ROOM: tba
EXAM (3 credits)
EXAM (3 credits)
13COMP
13COMP
Date: tba
Date: tba
Time: tba
Time: tba
DURATION: 2hrs to complete
DURATION: 2hrs to complete
ROOM: tba
ROOM: tba
what exactly is computer vision?
what exactly is computer vision?
Computer Vision image recognition and Deep Learning, helping the dream of smart cities become a reality.
Computer Vision image recognition and Deep Learning, helping the dream of smart cities become a reality.
From: TechSee
Computer vision is the field of computer science that focuses on replicating parts of the complexity of the human vision system and enabling computers to identify and process objects in images and videos in the same way that humans do.
Computer vision is the field of computer science that focuses on replicating parts of the complexity of the human vision system and enabling computers to identify and process objects in images and videos in the same way that humans do.
Until recently, computer vision only worked in limited capacity.
Until recently, computer vision only worked in limited capacity.
By 2022, the computer vision and hardware market is expected to reach $48.6 billion
By 2022, the computer vision and hardware market is expected to reach $48.6 billion
Computer vision is a field of computer science that works on enabling computers to see, identify and process images in the same way that human vision does, and then provide appropriate output. It is like imparting human intelligence and instincts to a computer. In reality though, it is a difficult task to enable computers to recognize images of different objects.
Computer vision is a field of computer science that works on enabling computers to see, identify and process images in the same way that human vision does, and then provide appropriate output. It is like imparting human intelligence and instincts to a computer. In reality though, it is a difficult task to enable computers to recognize images of different objects.
Computer vision is closely linked with artificial intelligence, as the computer must interpret what it sees, and then perform appropriate analysis or act accordingly.
Computer vision is closely linked with artificial intelligence, as the computer must interpret what it sees, and then perform appropriate analysis or act accordingly.
Useful links
Useful links
How Artificial Intelligence, Machine Learning & Deep Learning fit together
How Artificial Intelligence, Machine Learning & Deep Learning fit together
Artificial intelligence is a science like mathematics or biology. It studies ways to build intelligent programs and machines that can creatively solve problems, which has always been considered a human prerogative.
Machine learning is a subset of artificial intelligence (AI) that provides systems the ability to automatically learn and improve from experience without being explicitly programmed. In ML, there are different algorithms (e.g. neural networks) that help to solve problems.
Deep learning, or deep neural learning, is a subset of machine learning, which uses the neural networks to analyze different factors with a structure that is similar to the human neural system.
Click on one of the logos below to see more.
Click on one of the logos below to see more.
You will see they have slightly different views. It doesn't matter which model you select to run with, as long as you can clearly describe it.
You will see they have slightly different views. It doesn't matter which model you select to run with, as long as you can clearly describe it.
a good explanation of AI/ML/DL
a good explanation of AI/ML/DL
Video tutorials
Video tutorials
Start with the basics
Start with the basics
mentions:
kernals
convolution
Viola-Jones face detection
Convolution Neural Networks (CNN)
Facial Recognition
Facial Recognition
Facial Recognition uses nodal points (gap between the eyes, eye socket depth, etc) to recognise a face.
This is called a faceprint.
However, when the camera is at different angles this becomes difficult.
APIE (Aging Pose Illumination Emotion) allows computers to get around this issue using deepface. This process requires deep learning.
There is also facelt Argus which looks at your skin using surface texture analysis.
Computer vision vs Image Processing
Computer vision vs Image Processing
Machine learning explained in 5 levels of difficulty
Machine learning explained in 5 levels of difficulty
Machine learning
Machine learning
Only watch the first 3min
What is Deep Learning?
What is Deep Learning?
3:30 - 4:30 Some key aspects of Computer Vision.
So what is a neural network?
So what is a neural network?
Ever wondered what a computing neuron is?
OK, so it good upto just after 9min, then math occurs.....
Convolutional Neural Networks (CNN)
Convolutional Neural Networks (CNN)
"You have some input - like an image, and filter it using a convolution operation and repeat that process a number of times to learn something interesting about that image, some interesting features and then you make a classification decision based on it"
CNNs are able to detect patterns in images
Using machine learning and big data
Using machine learning and big data
5:50 No one tells a child how to see, especially in the early years. They learn it by seeing millions of images; by age 3 a child has seen hundred of millions of images.
"Instead of focusing solely on better and better algorithms, give the algorithms the kind of training data the child was given through experiences in both quantity and quality"
So the idea was to train the algorithms by downloading nearly a billion images of the internet and have people (almost 50,000 workers from 167 counties) to clean, sort and label the billion images.
These images became the training data for the neural networks.
How computers learn to recognize objects instantly
How computers learn to recognize objects instantly
Generative Adversarial Network (GAN)
Generative Adversarial Network (GAN)
Note: this video is about computers GENERATING something as opposed to computers "seeing" something.
A Generative Adversarial Network (GAN) is a class of machine learning frameworks designed by Ian Goodfellow and his colleagues in 2014.
Two neural networks contest with each other in a game (in the form of a zero-sum game, where one agent's gain is another agent's loss).
Moral dilemmas of computer vision
Moral dilemmas of computer vision
Legal problems in the UK over Police use of surveillance cameras
Legal problems in the UK over Police use of surveillance cameras
Difference between CPU and GPU
Difference between CPU and GPU
To understand the difference, we take a classic analogy which explains the difference intuitively.
Suppose you have to transfer goods from one place to the other. You have an option to choose between a Ferrari and a freight truck.
Ferrari would be extremely fast and would help you transfer a batch of goods in no time. But the amount of goods you can carry is small, and usage of fuel would be very high.
A freight truck would be slow and would take a lot of time to transfer goods. But the amount of goods it can carry is larger in comparison to Ferrari. Also, it is more fuel efficient so usage is lower.
So which would you chose for your work?
Obviously, you would first see what the task is; if you have to pick up your girlfriend urgently, you would definitely choose a Ferrari over a freight truck. But if you are moving your home, you would use a freight truck to transfer the furniture.
Machine Learning vs Deep Learning
Machine Learning vs Deep Learning
Machine Learning:
“A computer program is said to learn from experience E with respect to some class of tasks T and performance measure P if its performance at tasks in T, as measured by P, improves with experience E ”
Did that sound puzzling or confusing? click on Analytics Vidhya to find out more.
key aspects of computer vision
key aspects of computer vision
They want an overview, but remember, you may want to write in detail about these in question 2 & 3, so leave some details for later.
They want an overview, but remember, you may want to write in detail about these in question 2 & 3, so leave some details for later.
For 2022, questions may cover: noise, thresholding, edge detection, image processing for computer vision, applications.
For 2022, questions may cover: noise, thresholding, edge detection, image processing for computer vision, applications.
You could describe what Computer Vision is and then describe at least TWO key aspects of Computer Vision.
You could describe what Computer Vision is and then describe at least TWO key aspects of Computer Vision.
I think it useful to also include something on societal issues like:
I think it useful to also include something on societal issues like:
societies are beginning to talk about or implement regulation around computer vision.
shortage of computer vision specialists
potential for job losses.
You could also mention that Computer Vision is still developing.
You could also mention that Computer Vision is still developing.
Some suggestions:
Some suggestions:
Neural networks - training a neural network requires:
a massive amount of data to train with.
a lot of computational power - many Graphical Processing Units (GPU).
time.
a complex algorithm
Once "trained", using Computer Vision requires sufficient computer power to identify the object in a timely manner
OR
OR
Computer Vision required:
cheaper GPU due to mass production required to meet the gaming market
big data to enable the learning
better algorithms
OR
OR
From: heartbeat
Computer Vision could use:
image classification
2007 saw the use of machine learning
relevant algorithms OR other mechanisms behind Computer Vision.
relevant algorithms OR other mechanisms behind Computer Vision.
They want to see some detail on how it works.
They want to see some detail on how it works.
Some mechanisms you could write about:
Some mechanisms you could write about:
Viola Jones System.
Viola Jones System.
kernals.
kernals.
Convolutional Neural Networks, the videos above cover this.
Convolutional Neural Networks, the videos above cover this.
Hardware requirements - GPUs.
Hardware requirements - GPUs.
how is Computer Vision used?
how is Computer Vision used?
they want some detail on how it is implemented.
they want some detail on how it is implemented.
Augmented Reality & Mixed Reality
Augmented Reality & Mixed Reality
Face recognition: Snapchat and Facebook use face-detection algorithms to apply filters and recognize you in pictures.
Image retrieval: Google Images uses content-based queries to search relevant images. The algorithms analyze the content in the query image and return results based on best-matched content.
Gaming and controls: A great commercial product in gaming that uses stereo vision is Microsoft Kinect.
Surveillance: Surveillance cameras are ubiquitous at public locations and are used to detect suspicious behaviors.
Biometrics: Fingerprint, iris and face matching remains some common methods in biometric identification.
Smart cars: Vision remains the main source of information to detect traffic signs and lights and other visual features.
Beware of the hype surrounding new technology.
Beware of the hype surrounding new technology.
"When new technologies make bold promises, how do you discern the hype from what’s commercially viable? And when will such claims pay off, if at all? "
key problems or issues related to Computer Vision
key problems or issues related to Computer Vision
explain at least two key problems or issues.
explain at least two key problems or issues.
Issue with algorithms , solved by Convolutional Neural Networks
Issue with algorithms , solved by Convolutional Neural Networks
Requirement for hardware improvement - GPUs
Requirement for hardware improvement - GPUs
Lack of specialists
Lack of specialists
Need for regular monitoring
Need for regular monitoring
Facial recognition:
Facial recognition:
privacy issues
societies desire for regulation of its use.
racial issues
Necessity of specialists: there is a huge necessity of specialist related to the field of Machine Learning and Artificial Intelligence. A professional that knows how those devices work and take full advantage of Computer Vision. Also, the person can repair them when necessary. There are a lot of work opportunities after doing a Master in Artificial Intelligences. However, companies still wait for those specialists.
Spoiling: eliminate the human factor may be good in some cases. But when the machine or device fails, it doesn’t announce or anticipate that problem. Whereas a human person can tell in advance when the person won’t come.
Failing in image processing: when the device fails because of a virus or other software issues, it is highly probable that Computer Vision and image processing will fail. But if we do not solve the problem, the functions of the device can dissapear. It can froze the entire production in the case of warehouses.
No technology is free from flaws. And the same applies to computer vision systems. Let’s now look at a few limitations that the technology inherently has:
Lack of specialists - Computer vision technology involves the use of AI and ML. To train a computer vision system powered by AI and ML, companies need to have a team of professionals with technical expertise. Without them, building a system that can analyze and process the possible surrounding details is not possible.
Need for regular monitoring - What if a computer vision system breaks down or has a technical glitch? To ensure that doesn’t happen, companies have to get a dedicated team onboard for regular monitoring and evaluation.
Despite their current limitations, computer vision systems can bring companies immense opportunities to increase revenue streams, meet productivity goals, and streamline work processes. However, we have barely just scratched the surface of computer vision capabilities. The future is yet to be seen.
technical capabilities & limitations of Computer Vision relate to humans
technical capabilities & limitations of Computer Vision relate to humans
detailed explanation of both technical capabilities & limitations.
detailed explanation of both technical capabilities & limitations.
need to relate it to humans
need to relate it to humans
Advantages
Advantages
The use of Computer Imagining grows rapidly thanks to the discovery of advantages for industries. There are five main advantages of computer vision:
Process in a simpler and faster way: it allows the clients and industries to quality check products. Also, it gives them access to their products. It’s possible thanks to the existence of Computer Vision in fast computers.
Reliability: computers and cameras don’t have the human factor of tiredness, which is eliminated in them. The efficiency is usually the same, it doesn’t depend on external factors such as illness or sentimental status.
Accuracy: the precision of Computer Imagining, and Computer Vision will ensure a better accuracy on the final product.
A wide range of use: We can see the same computer system in several different fields and activities. Also, in factories with warehouse tracking and shipping of supplies, and in the medical industry through scanned images, among other multiple options.
The reduction of costs: time and error rate are reduced in the process of Computer Imagining. It reduces the cost of hire and train special staff to do the activities that computers will do as hundreds of workers.
Disadvantages
Disadvantages
Despite all the advantages of computer vision thanks to the capacity of Machine Learning, we have to consider some disadvantages:
Necessity of specialists: there is a huge necessity of specialist related to the field of Machine Learning and Artificial Intelligence. A professional that knows how those devices work and take full advantage of Computer Vision. Also, the person can repair them when necessary. There are a lot of work opportunities after doing a Master in Artificial Intelligences. However, companies still wait for those specialists.
Spoiling: eliminate the human factor may be good in some cases. But when the machine or device fails, it doesn’t announce or anticipate that problem. Whereas a human person can tell in advance when the person won’t come.
Failing in image processing: when the device fails because of a virus or other software issues, it is highly probable that Computer Vision and image processing will fail. But if we do not solve the problem, the functions of the device can dissapear. It can froze the entire production in the case of warehouses.
Advantages
Advantages
Faster and simpler process - Computer vision systems can carry out monotonous, repetitive tasks at a faster rate, making the entire process simpler.
Accurate outcome - It's no secret that machines never make any mistake. Likewise, computer vision systems with image-processing capabilities will commit zero mistakes, unlike humans. Ultimately, products or services provided will not only be quick but also of high quality.
Cost-reduction - With machines taking up responsibilities of performing cumbersome tasks, errors will be minimized, leaving no room for faulty products or services. As a result, companies can save a lot of money that would be otherwise spent on fixing flawed processes and products.
Limitations
Limitations
No technology is free from flaws. And the same applies to computer vision systems. Let’s now look at a few limitations that the technology inherently has:
Lack of specialists - Computer vision technology involves the use of AI and ML. To train a computer vision system powered by AI and ML, companies need to have a team of professionals with technical expertise. Without them, building a system that can analyze and process the possible surrounding details is not possible.
Need for regular monitoring - What if a computer vision system breaks down or has a technical glitch? To ensure that doesn’t happen, companies have to get a dedicated team onboard for regular monitoring and evaluation.
Despite their current limitations, computer vision systems can bring companies immense opportunities to increase revenue streams, meet productivity goals, and streamline work processes. However, we have barely just scratched the surface of computer vision capabilities. The future is yet to be seen.
different perspectives on Computer Vision
different perspectives on Computer Vision
compare & contrast at least two perspectives.
compare & contrast at least two perspectives.
Computer vision is good:
Computer vision is good:
helps police identify criminal's movement.
helps police identify criminal's movement.
self driving transport
self driving transport
helps Government identify terrorists and Foreign Government agents
helps Government identify terrorists and Foreign Government agents
healthcare
healthcare
make shopping seamless
make shopping seamless
Computer Vision is bad:
Computer Vision is bad:
issues around misuse by Government agencies
issues around misuse by Government agencies
missuses by Corporations
missuses by Corporations
job losses
job losses
military uses
military uses
The privacy concerns relating to facial recognition have received widespread attention. In 2019, San Francisco and Oakland, California, and a trio of Massachusetts communities banned facial recognition tools for use in law enforcement. Portland, Oregon, plans to do the same.
The privacy concerns relating to facial recognition have received widespread attention. In 2019, San Francisco and Oakland, California, and a trio of Massachusetts communities banned facial recognition tools for use in law enforcement. Portland, Oregon, plans to do the same.
conclusions about Computer vision
conclusions about Computer vision
Should link back to your section on pros and cons of computer vision.
Should link back to your section on pros and cons of computer vision.
explore less-obvious implications:
explore less-obvious implications:
having a machine make moral decisions. EG: a self driving car is presented with a choice of who to save, which one will it be?
computer vision is the missing building block to "enable" computers. EG: computer's are able to process vast amounts of data very quickly, but have not been able to see, limiting what computers can do.
jobs losses in some less obvious fields. EG: everyone knows about taxi drivers being replaced by self driving cars, but what about shop assistants being replaced by Amazon Go? And what about other jobs
a computer vision system may pass Government certification to be able to be sold, but what checks will be done to ensure the system is still functioning correctly. EG: WOF for vehicles.
From: Soulpage Behavior determination helps in determining the mood and engagement level of employees while performing tasks so that organizations can re-work on their operations for better outputs. Object recognition, action recognition, crowd analysis, etc., are other prominent areas where computer vision is used to enhance employee performance.
face recognition to authorise online shopping or online banking.
potential improvements to computer vision:
potential improvements to computer vision:
?
suggest innovative and imaginative wider uses:
suggest innovative and imaginative wider uses:
dangerous jobs. EG: mining, diving.
certification of equipment by machine. EG: vehicle WOF checking to carried out by computer controlled devices with computer vision. OR air plane certification, etc.
Some websites on writing conclusions:
Some websites on writing conclusions:
Page updated
Report abuse