Quiz Space

Deep Learning Practice · End Term · 10 May 2026 · January 2026 term

Question 12: Regarding the U-Net architecture, which is widely used f…

Question 12

+5 marksOne or more correct options

Regarding the U-Net architecture, which is widely used for tasks like medical image segmentation and can be adapted for depth estimation:

Select all that apply.

  1. A

    It features a symmetric architecture consisting of a contracting path (encoder) and an expansive path (decoder).

  2. B

    Skip connections concatenate high-resolution features from the contracting path directly to the up sampled features in the expansive path.

  3. C

    The expansive path uses transposed convolutions (or up-convolutions) to increase the spatial resolution of the feature maps.

  4. D

    It relies on Global Average Pooling at every layer to ensure that spatial information is discarded in favor of global context.

Show answer

Correct answers

  • A

    It features a symmetric architecture consisting of a contracting path (encoder) and an expansive path (decoder).

  • B

    Skip connections concatenate high-resolution features from the contracting path directly to the up sampled features in the expansive path.

  • C

    The expansive path uses transposed convolutions (or up-convolutions) to increase the spatial resolution of the feature maps.

See the whole paperTake it as a mock test

Question 12 of 22 in the IIT Madras BS Deep Learning Practice (Deep Learning Practice) End Term paper sat on 10 May 2026, in the January 2026 term (Deep Learning Practice 06 May 26). It carries 5 marks.

More questions from this paper

  1. Q1Based on the YAML configuration file, what is the total number of object classes the model is trained to detect?
  2. Q2What is the x-coordinate (in pixels) of the center of the bounding box?
  3. Q3What is the x-coordinate (in pixels) of the top-left corner of the bounding box?
  4. Q4Consider a convolutional neural network with an input image of size 84 × 84 × 3. The architecture consists of the follo…
  5. Q5A bottleneck layer in a ResNet receives an input feature map of dimensions 14×14×1024. We apply a 1×1 convolution layer…
  6. Q6A modified YOLO-style architecture divides an image into a 13 × 13 grid (S = 13). If the model is designed to predict B…
  7. Q7In the VGG16 architecture, consider the second convolutional layer of the first block. It takes an input with 64 channe…
  8. Q8When fine-tuning a large vision model (e.g., a pre-trained ResNet or Vision Transformer) for a specific competition dat…
  9. Q9When training a deep learning model for a vision competition (such as a ”Cat vs Dog” classifier or a YOLO detector), wh…
  10. Q10Consider the following Python code snippet using the Ultralytics YOLOv8 API:\ from ultralytics import YOLO model = YOLO…
  11. Q11Which of the following loss components are typically used in SRGAN (Super-Resolution GAN) to achieve photo-realistic re…
  12. Q13Consider the following code:\ model = torchvision.models.detection.fasterrcnn_resnet50_fpn(pretrained=True) images = [t…
  13. Q14Which of the following statements correctly describe the characteristics and motivations of the Inception (GoogLeNet) a…
  14. Q15In the context of Convolutional Neural Networks (CNNs), what is the primary mechanism that allows the network to handle…
  15. Q16The first layer of the VGG16 model is a Conv2d with kernel = 3, stride = 1, padding = 1.\ Consider the following Python…
  16. Q17In the ResNet (Residual Network) architecture, 1 × 1 convolutions are frequently used in the ”bottleneck” building bloc…
  17. Q18In the YOLO (You Only Look Once) framework, if multiple bounding boxes are predicted by a single grid cell for the same…
  18. Q19Consider the following code snippet using a YOLO-based object detection model:\ from ultralytics import YOLO\ model = Y…
  19. Q20In the SRResNet and SRGAN architectures for image super-resolution, which specialized layer is used to upsample the fea…
  20. Q21In the Fast R-CNN architecture, which component is responsible for extracting a fixed-size feature vector from a shared…
  21. Q22In the GoogLeNet (Inception v1) architecture, 1 × 1 convolutions (bottleneck layers) are applied before larger 3 × 3 an…