CNN CNN - Convolutional Neural Networks Semantic Segmentation Questions and Answers 1 — Questions and Answers
Question 1: What is the goal of semantic segmentation in computer vision?
- Detecting bounding boxes around objects
- Assigning a class label to every pixel in an image (Correct answer)
- Identifying unique instances of each object
- Generating captions for images
Correct answer: Assigning a class label to every pixel in an image
Semantic segmentation classifies each pixel into a predefined category, producing a dense prediction map that delineates different regions of the image.
Question 2: What key innovation did Fully Convolutional Networks (FCN) introduce for semantic segmentation?
- Using fully connected layers for dense prediction
- Replacing fully connected layers with convolutional layers to enable end-to-end pixel-wise prediction (Correct answer)
- Adding attention mechanisms
- Using recurrent layers for sequential processing
Correct answer: Replacing fully connected layers with convolutional layers to enable end-to-end pixel-wise prediction
FCN replaced the fully connected layers of classification networks with convolutional layers, allowing the network to accept arbitrary-sized inputs and produce spatial output maps.
Question 3: What architecture is U-Net primarily designed for, and what is its signature structural feature?
- Object detection with anchor boxes
- Biomedical image segmentation with a symmetric encoder-decoder and skip connections (Correct answer)
- Image classification with residual blocks
- Video segmentation with temporal convolutions
Correct answer: Biomedical image segmentation with a symmetric encoder-decoder and skip connections
U-Net is a symmetric encoder-decoder network with skip connections that directly concatenate feature maps from the encoder to the corresponding decoder level, preserving fine-grained spatial detail.
Question 4: What is the role of skip connections in the U-Net architecture?
- They skip residual computations to speed up training
- They pass high-resolution feature maps from the encoder to the decoder to recover spatial detail lost during downsampling (Correct answer)
- They skip low-performing layers during inference
- They connect non-adjacent layers for gradient flow only
Correct answer: They pass high-resolution feature maps from the encoder to the decoder to recover spatial detail lost during downsampling
Skip connections concatenate encoder feature maps with decoder feature maps at matching resolutions, allowing the decoder to recover fine spatial details that are lost during pooling.
Question 5: What is the fundamental difference between semantic segmentation and instance segmentation?
- Semantic segmentation is faster
- Semantic segmentation assigns the same label to all pixels of a class, while instance segmentation distinguishes individual object instances (Correct answer)
- Instance segmentation works only on videos
- Semantic segmentation requires depth data
Correct answer: Semantic segmentation assigns the same label to all pixels of a class, while instance segmentation distinguishes individual object instances
In semantic segmentation all pixels of the same class share one label, whereas instance segmentation distinguishes separate instances so two cars get different masks.
Question 6: What is the encoder-decoder architecture in CNN-based segmentation networks?
- An autoencoder used for image compression
- A network where the encoder downsamples the input to a bottleneck and the decoder upsamples it back to the original resolution for pixel prediction (Correct answer)
- A two-stage classifier followed by a regressor
- A GAN-based structure for generating masks
Correct answer: A network where the encoder downsamples the input to a bottleneck and the decoder upsamples it back to the original resolution for pixel prediction
The encoder extracts hierarchical features while reducing spatial resolution, and the decoder gradually upsamples these features back to the input resolution to produce the segmentation map.
CNN CNN - Convolutional Neural Networks Semantic Segmentation Questions and Answers 1
What is the goal of semantic segmentation in computer vision?