Autoencoder for images in Computer Vision - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Model Choice

intermediate

Choosing the right architecture for an image autoencoder

You want to build an autoencoder to compress and reconstruct 28x28 grayscale images. Which model architecture is best suited for this task?

A A convolutional neural network with convolutional layers followed by a bottleneck and deconvolutional layers

B A fully connected neural network with input size 784 and a bottleneck layer of size 32

C A recurrent neural network with LSTM layers to process the image pixels sequentially

D A decision tree model trained to predict pixel values from other pixels

💡 Hint

Think about which model type can capture spatial patterns in images effectively.

Convolutional autoencoders use convolutional layers to capture spatial features in images, making them better for image compression and reconstruction than fully connected or sequential models.

❓ Predict Output

intermediate

Output shape after encoding in a convolutional autoencoder

Given the following encoder code for 28x28 grayscale images, what is the shape of the encoded output?

Computer Vision

import torch
import torch.nn as nn

class Encoder(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 16, 3, stride=2, padding=1)  # output: 16 x 14 x 14
        self.conv2 = nn.Conv2d(16, 8, 3, stride=2, padding=1)  # output: 8 x 7 x 7
    def forward(self, x):
        x = torch.relu(self.conv1(x))
        x = torch.relu(self.conv2(x))
        return x

encoder = Encoder()
input_tensor = torch.randn(1, 1, 28, 28)
encoded = encoder(input_tensor)
print(encoded.shape)

A torch.Size([1, 16, 14, 14])

B torch.Size([1, 8, 7, 7])

C torch.Size([1, 8, 14, 14])

D torch.Size([1, 1, 28, 28])

💡 Hint

Calculate output size after each convolution using stride and padding.

First conv reduces 28x28 to 14x14 with 16 channels, second conv reduces 14x14 to 7x7 with 8 channels, so output shape is [1, 8, 7, 7].

❓ Hyperparameter

advanced

Choosing the right latent dimension size in an autoencoder

You are training an autoencoder on 64x64 RGB images. What is the effect of choosing a very small latent dimension size?

A The model will compress images too much, causing blurry or poor reconstructions

B The model will overfit and memorize training images exactly

C The model will run faster but use more memory

D The model will reconstruct images perfectly with no loss

💡 Hint

Think about what happens if the compressed representation is too small to hold enough information.

A very small latent space forces the model to lose important details, leading to blurry or inaccurate reconstructions.

❓ Metrics

advanced

Evaluating autoencoder reconstruction quality

Which metric is best to measure how well an autoencoder reconstructs images?

A F1 score of reconstructed image pixels

B Accuracy of predicted class labels

C Cross-entropy loss between input and output pixels

D Mean Squared Error (MSE) between original and reconstructed images

💡 Hint

Consider a metric that measures pixel-wise difference between images.

MSE measures the average squared difference between original and reconstructed pixels, making it suitable for image reconstruction quality.

🔧 Debug

expert

Identifying the cause of poor autoencoder training

You trained an autoencoder on 32x32 grayscale images but the loss does not decrease and reconstructions are random noise. What is the most likely cause?

A The batch size is too large causing slow convergence

B The input images are not normalized before training

C The learning rate is too high causing unstable training

D The autoencoder has too many layers causing overfitting

💡 Hint

Think about what causes training loss to not improve and outputs to be random.

A very high learning rate can cause the model weights to update too aggressively, preventing loss from decreasing and producing noisy outputs.

Practice

1. What is the main purpose of an autoencoder when applied to images?

easy

A. To generate new images from random noise

B. To classify images into categories

C. To compress and then reconstruct the image data

D. To detect edges using filters

Solution

1. Step 1: Understand autoencoder function

   An autoencoder learns to compress input data into a smaller representation and then reconstruct it back.

2. Step 2: Apply to images

   For images, this means compressing image pixels and reconstructing the image to learn important features.

3. Final Answer:

   To compress and then reconstruct the image data -> Option C

4. Quick Check:

   Autoencoder = compression + reconstruction [OK]

Hint: Autoencoder = compress + rebuild images [OK]

Common Mistakes:

• Confusing autoencoder with classifier
• Thinking autoencoder generates new images
• Mixing edge detection with autoencoding

2. Which of the following is the correct way to define the encoder part of a convolutional autoencoder in Python using Keras?

easy

A. encoder = Sequential([LSTM(32, input_shape=(28,28,1))])

B. encoder = Sequential([Dense(32, activation='relu', input_shape=(28,28,1))])

C. encoder = Sequential([Conv1D(32, 3, activation='relu', input_shape=(28,28,1))])

D. encoder = Sequential([Conv2D(32, (3,3), activation='relu', input_shape=(28,28,1)), MaxPooling2D((2,2))])

Solution

1. Step 1: Identify convolutional layers for images

   Conv2D layers are used for 2D image data, with filters and kernel size.

2. Step 2: Check correct syntax and input shape

   encoder = Sequential([Conv2D(32, (3,3), activation='relu', input_shape=(28,28,1)), MaxPooling2D((2,2))]) uses Conv2D with activation and input shape for grayscale images (28x28x1), plus pooling for downsampling.

3. Final Answer:

   encoder = Sequential([Conv2D(32, (3,3), activation='relu', input_shape=(28,28,1)), MaxPooling2D((2,2))]) -> Option D

4. Quick Check:

   Conv2D + MaxPooling2D = convolutional encoder [OK]

Hint: Use Conv2D for images, not Dense or Conv1D [OK]

Common Mistakes:

• Using Dense layer instead of Conv2D for images
• Using Conv1D which is for sequences, not images
• Using LSTM which is for sequences, not images

3. Given this simple convolutional autoencoder code snippet, what will be the shape of the encoded output?

from tensorflow.keras.layers import Input, Conv2D, MaxPooling2D
from tensorflow.keras.models import Model
input_img = Input(shape=(28, 28, 1))
x = Conv2D(16, (3, 3), activation='relu', padding='same')(input_img)
x = MaxPooling2D((2, 2), padding='same')(x)
encoded = Conv2D(8, (3, 3), activation='relu', padding='same')(x)
model = Model(input_img, encoded)

medium

A. (14, 14, 8)

B. (28, 28, 8)

C. (7, 7, 8)

D. (14, 14, 16)

Solution

1. Step 1: Analyze input and first Conv2D layer

   Input shape is (28,28,1). First Conv2D with padding='same' keeps spatial size (28,28), output channels 16.

2. Step 2: Apply MaxPooling2D with pool size (2,2) and padding='same'

   This halves spatial dimensions: (28,28) -> (14,14), channels remain 16.

3. Step 3: Apply second Conv2D with 8 filters and padding='same'

   Spatial size stays (14,14), channels become 8.

4. Final Answer:

   (14, 14, 8) -> Option A

5. Quick Check:

   Pooling halves size, Conv2D changes channels [OK]

Hint: Pooling halves size, Conv2D changes channels [OK]

Common Mistakes:

• Ignoring padding='same' effect on size
• Assuming Conv2D changes spatial size
• Confusing channels with spatial dimensions

4. You wrote this decoder code for your convolutional autoencoder but get an error during training:

decoder = Sequential([
  Conv2D(8, (3,3), activation='relu', padding='same', input_shape=(14,14,16)),
  UpSampling2D((2,2)),
  Conv2D(1, (3,3), activation='sigmoid', padding='same')
])

What is the likely cause of the error?

medium

A. Missing UpSampling2D layer before the last Conv2D

B. Input shape to decoder does not match encoded output shape

C. Activation 'sigmoid' is not allowed in Conv2D

D. Conv2D filters must be 16, not 8 or 1

Solution

1. Step 1: Check encoded output shape

   The encoder output shape is (14,14,8) after the final Conv2D layer, but decoder input shape=(14,14,16).

2. Step 2: Verify decoder input shape matches encoded output

   Decoder input_shape=(14,14,16) does not match encoded output (14,14,8), causing shape mismatch in channels during training.

3. Step 3: Identify common error cause

   Often error arises if decoder input shape does not match encoder output shape exactly, causing shape mismatch during training.

4. Final Answer:

   Input shape to decoder does not match encoded output shape -> Option B

5. Quick Check:

   Decoder input shape must match encoder output [OK]

Hint: Decoder input shape must match encoder output shape [OK]

Common Mistakes:

• Thinking activation 'sigmoid' is invalid
• Assuming filters must be 16 always
• Ignoring input shape mismatch errors

5. You want to use a convolutional autoencoder to remove noise from grayscale images of size 64x64. Which architecture choice is best to preserve spatial details while reducing noise?

hard

A. Use multiple Conv2D layers with small (3x3) kernels and padding='same', plus MaxPooling2D for downsampling

B. Use Dense layers only to compress the flattened image vector

C. Use Conv1D layers with large kernels to capture noise patterns

D. Use a single Conv2D layer without pooling or upsampling

Solution

1. Step 1: Understand noise removal with autoencoders

   Noise removal requires learning spatial features and reconstructing clean images, so preserving spatial info is key.

2. Step 2: Evaluate architecture choices

   Multiple Conv2D layers with small kernels and padding='same' keep spatial size and details; MaxPooling2D reduces size to learn compressed features.

3. Step 3: Reject other options

   Dense layers lose spatial info; Conv1D is for sequences; single Conv2D without pooling won't compress well.

4. Final Answer:

   Use multiple Conv2D layers with small (3x3) kernels and padding='same', plus MaxPooling2D for downsampling -> Option A

5. Quick Check:

   Conv2D + pooling preserves spatial details for denoising [OK]

Hint: Use Conv2D + pooling to keep spatial info and reduce noise [OK]

Common Mistakes:

• Using Dense layers losing image structure
• Using Conv1D which is for 1D data
• Skipping pooling reduces compression ability