PCV-SLM3

What is the primary goal of object recognition in the field of computer vision?

• Creating realistic images

• Identifying objects in images and videos

• Generating 3D models

• Enhancing image resolution

Before the era of deep learning, what was a state-of-the-art method for object detection using machine learning?

• Convolution Neural Network (CNN)

• Bag of features model

• Viola-Jones algorithm

• HOG and SVM model

What advantage does deep learning have over traditional machine learning in object recognition?

• Faster computation

• No need for feature extraction

• Better accuracy with small datasets

• Simplicity in implementation

In image classification, what does the algorithm output for a given input image?

• Bounding box

• Class label

• Probability distribution

• Regression values

What does an object localization algorithm output for a given input image?

• Probability distribution

• Class label

• Bounding box

• Segmentation mask

How do object detection algorithms combine image classification and object localization?

• By using semantic segmentation

• By generating pixel-wise masks

• By producing one or more bounding boxes with class labels

• By applying HOG features

What challenge is associated with the bounding boxes generated in object detection?

• They are always circular

• They cannot determine object shapes with curvature

• They only provide class labels

• They are computationally expensive

What distinguishes image segmentation from object detection?

• Image segmentation uses bounding boxes.

• Object detection outputs pixel-wise masks.

• Image segmentation doesn't consider object shapes.

• Object detection focuses on class labels only.

Which segmentation approach is more granular and helps determine the shape of each object in an image?

• Instance Segmentation

• Object Detection

• Object Localization

• Image Classification

What is the primary goal of Instance Segmentation?

• Labeling each pixel in the image with different colors

• Identifying the boundaries of objects

• Detecting diseases in medical images

• Classifying images into categories