Scribd
Upload
13 views

Mathematics for Computer Vision

The document outlines essential mathematical concepts for computer vision, focusing on Linear Algebra, Calculus, and Probability and Statistics. Key topics include matrix operations, derivatives for edge detection, and various probability distributions such as Bernoulli and Normal distributions. It also includes assignments related to gradient computation and random permutations using Python.

Uploaded by

gurjibecha88
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
13 views

Mathematics for Computer Vision

The document outlines essential mathematical concepts for computer vision, focusing on Linear Algebra, Calculus, and Probability and Statistics. Key topics include matrix operations, derivatives for edge detection, and various probability distributions such as Bernoulli and Normal distributions. It also includes assignments related to gradient computation and random permutations using Python.

Uploaded by

gurjibecha88
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 14

Mathematics for

computer vision

2
Linear Algebra (matrices,
transformations)
Learning
Outline
Calculus (gradients,
optimization)

Probability and
Statistics
Linear Algebra
Computer Vision deals with knowledge Key Linear Algebra
extraction from images. And for the Concepts
extraction, core concepts of Linear Algebra
(LA) are used. We have discussed earlier the Ø Determinants, inverse
concepts of Linear algebra are Vectors, Ø Singular value
Matrices and Tensors. decomposition (SVD)
Ø Singular matrices
Ø Orthogonality / Cross &
Dot
Ø Eigen values & Eigen
vectors
Ø Convolution math

6
Understanding Calculus

Calculus with respect to computer


vision is used to analyze the rate of
Calculus
change in pixel values across an
image to identify edges. This is
done using derivatives.
Automatic Differentiation in PyTorch
Click here to Access the
 Gradients using torch. Notebook (Matrices and
 Derivatives using torch Calculus) Notebook Reference
Click here to access
reference for Linear
algebra
Probability and Statistics

Uniform Bernoulli Binomial • Normal


Distribution Distribution Distribution Distribution

Poisson Multinomial Geometric Chi-


Squared
Distribution Distribution • Distribution
Distribution
Probabilities and Statistics
Uniform

The uniform distribution is one of


the simplest probability
distribution in statistics. It is a
continuous distribution between
0 and 1. It represents equal
likelyhood of all outcomes.

Bernoulli

This is a discrete probability


distribution for a random variable
that has only two possible
outcomes: success (1) or failure
(0). It is characterized by a single
parameter p, which represents the
probability of success.
Probabilities and Statistics
Binomial

Binomial distribution is a distribution


calculated by multiplying the probability
of success raised to the power of the
number of successes and the probability
of failure raised to the power of the
difference between the number of
successes and the number of trials.

Normal

This is also known as the gaussian


distribution described as a symmetrical
plot of data around it’s mean value,
where the width of the curve is defined
by the standard deviation.
Probabilities and Statistics
Poisson

This is a discrete probability distribution that


expresses the probability of a given number
of events occurring in a fixed interval of time.
It could also be defined as the calculation of
the likelihood of a certain number of events
happening in a fixed time or space, assuming
the events occur independently and at a
constant rate.

Multinomial

A probability distribution used to determine


the likelyhood of a certain set of outcomes.
Probabilities and Statistics
Geometric

It’s a discrete probability


distribution that describes the
chances of achieving success in a
series of independent trials, each
having two possible outcomes. This
helps measure the probability of
success after a given number of
trials.

Chi-Squared

The chi-squared distribution is one of


the most widely used probability
distributions in inferential statistics,
notably in hypothesis testing and in
construction of confidence intervals.
(Probabilities and
Statistics)
Click here to Access the
Notebook (Probabilities
and Statistics)
Notebook (Probabilities
and Statistics)

References

Probabilities
(https://pytorch.org/docs/s
table/distributions.html)
Assignment 1

MILESTONE

1 2 3 4 5

Compute the What is the Submit your


Open a gradient of gradient of answer as a
notebook f(x) = x^4 + f(x) when response
4x^3 + 10x^2 x=3
+ 3x + 11

START
Assignment 2

 Using uniform distribution and randperm, write a


Python function called customperm, that takes in
input (a,b,c) and outputs a tensor of random
permutations between 0 and c, with a dimension
(a,b).

 Try this, customperm(3,4,20)

You might also like