Scribd
Upload
8 views3 pages

Assignment 2

Uploaded by

Vemula Muni Karthik
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
8 views3 pages

Assignment 2

Uploaded by

Vemula Muni Karthik
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/ 3

Sri Sivasubramaniya Nadar College of Engineering, Kalavakkam – 603 110

(An Autonomous Institution, Affiliated to Anna University, Chennai)


Department of Information Technology
Assignment – 2

Degree & Branch B.Tech. Information Technology Semester VII


Subject Code & Name UIT2729– IMAGE PROCESSING AND COMPUTER VISION

Course Outcome:
CO1- Use the different techniques used for feature detection and matching.
CO2-Apply the different techniques used for image segmentation.
CO3-Understand depth estimation and 3-D reconstruction.
CO4-Apply the different types of recognition used in computer vision.
CO5- Apply the different deep learning models and networks used in computer vision.

K1: Remembering K2: Understanding K3: Applying K4: Analyzing K5: Evaluating

KL – Knowledge Level CO – Course Outcomes PI – Performance Indicator

Q. KL Questions CO
No

1 Consider the following input image. Apply the YOLO algorithm, split
the image into 3*3 grids, and use the bounding box method to locate
the objects in the image. Apply the Intersection over Union and Non-
Max Suppression.

CO4 &
K3
CO5

2 Sketch the training process of Siamese networks, exposing the process CO4 &
K3
of training with positive samples and negative samples. Use the triple CO5
loss function to train the following data set. Calculate the distance
calculation using the Triplet loss function.

Triplet loss-Training data set

Triplet loss-Training data set with positive sample

Triplet loss-Training data set with negative sample

3 K4 Compare a semantic segmentation model (e.g., FCN) with an CO4 &


instance segmentation model (e.g., Mask R-CNN) on the same CO5
dataset. How do they differ in terms of accuracy, memory usage, and
inference time?

4 K3 You are building a deep learning model to classify different stages of


growth for tomato plants (e.g., seedling, vegetative, flowering, and
CO4 &
fruit-bearing stages). You have collected a dataset of images of tomato
CO5
plants at each stage and would like to classify each image according to
these growth stages.

---------------------------------

You might also like