Exp1: VIVA QUESTIONS
1. What is the mathematical formula for the Gaussian distribution, and how does each parameter
μ and σ2 affect its shape?
2. In your MATLAB/Python function, how do you handle the computation of the Gaussian PDF
for a vector X instead of a single scalar value?
3. Explain the purpose of normalizing the Gaussian distribution and how the normalization
factor 1/√2𝜋𝜎2 is derived?
4. How does varying the mean (μ) and variance (σ2) in your plot help in understanding the
properties of the Gaussian distribution?
5. Describe the process of plotting the Gaussian distribution for different parameters in
MATLAB/Python. What key steps are involved, and how do you ensure accurate visualization?
Exp 2 VIVA QUESTIONS
1. Can you explain the main concept behind the gradient descent algorithm and how it helps
in optimizing functions?
2. What role does the learning rate (α) play in the gradient descent algorithm, and how does it
affect the convergence of the algorithm?
3. How do you handle the issue of local minima in gradient descent, especially in the context
of complex cost functions?
4. Describe the process of computing the gradient in the context of gradient descent. How do
you ensure that the gradient computation is accurate?
5. How would you modify the gradient descent algorithm to handle a scenario where the cost
function is very noisy or has high variance?
Exp3 VIVA QUESTIONS
1. What is the purpose of the cost function in linear regression?
2. What is the difference between simple linear regression and multiple linear regression?
3. Why do we use the gradient of the cost function in gradient descent?
4. What happens if the learning rate α\alphaα is too large or too small?
5. How do you know when to stop the gradient descent algorithm?
Exp 4 VIVA QUESTIONS:
1. Why is it important to compare the classification accuracy of different models like SVM,
ELM, and CNN?
2. How does SVM find the optimal hyperplane for classification?
3. What is an Extreme Learning Machine (ELM), and how does it differ from traditional neural
networks?
4.
5. How would you empirically compare the classification accuracy of SVM, ELM, and CNN on
a given dataset?
6. What factors influence the classification accuracy of SVM, ELM, and CNN?
Exp 5 VIVA QUESTIONS
1. How do you read and display an image using Python libraries like OpenCV or PIL?
2. What is the significance of resizing an image, and how would you implement it?
3. How would you convert a color image to grayscale, and why might you want to do this?
4. What is the purpose of blurring an image, and how can you implement a Gaussian blur?
5. How do you perform edge detection on an image, and what role does it play in image processing?
Exp 6 VIVA QUESTIONS
1. Explain the basic geometric transformations and their matrix representations.
2. What is the difference between affine and non-affine transformations?
3. How is homogeneous coordinates used in geometric transformations?
4. Describe the process of rotating a point around an arbitrary axis in 3D space.
5. How do you handle transformations involving scaling, especially when non-uniform scaling
is applied?
Exp 7 VIVA QUESTIONS
1. Describe the process of applying a scaling transformation to a 2D image.
2. Explain how translation is implemented in 3D graphics.
3. What are the key steps involved in performing a rotation transformation around an arbitrary
axis in 3D space?
4. Discuss the role of homogeneous coordinates in geometric transformations.
5. How do you implement a perspective transformation, and what challenges might arise during
its application?
Exp 8 VIVA QUESTIONS
1. What steps are involved in the camera calibration process, and why is each step important?
2. How does the choice of calibration pattern (e.g., checkerboard vs. circle grid) impact the
calibration process?
3. Explain the concept of reprojection error in the context of camera calibration.
4. Discuss the process of estimating lens distortion parameters during camera calibration.
5. How do you handle dynamic scenes or moving objects during the calibration process?
Exp 9 VIVA QUESTIONS
1. 1. What is the fundamental matrix, and how does it relate corresponding points between two images?
2. Explain the process of feature detection and matching in the context of computing the fundamental matrix.
3. Describe the eight-point algorithm for estimating the fundamental matrix.
4. Why is it necessary to enforce the rank-2 constraint on the fundamental matrix, and how is this typically achieved?
Exp 10 VIVA QUESTIONS
1. What are the main stages involved in hand gesture recognition?
2. Explain the concept of human pose detection and how it is typically represented in computer
vision systems.
3. How do deep learning models like OpenPose or PoseNet contribute to human pose detection?
4. Describe how hand gesture recognition can be integrated with human pose detection to
enhance human-computer interaction.
5. What are the common challenges in real-time hand gesture recognition, and how can they be
addressed?
EXP 11 VIVA QUESTIONS
1. Explain the difference between face detection and face recognition.
2. What is the role of the Haar Cascade Classifier in face detection?
3. Describe the process of face encoding in the context of face recognition.
4. How would you implement real-time face recognition using a webcam with Python and
OpenCV?
5. What are the benefits of using the face_recognition library over just using OpenCV for face
recognition?
