Why architecture design impacts performance in Computer Vision - Why Metrics Matter

Metrics & Evaluation - Why architecture design impacts performance

Which metric matters and WHY

In computer vision, architecture design affects how well a model learns and predicts. Key metrics include accuracy for overall correctness, precision and recall for class-specific performance, and F1 score to balance precision and recall. These metrics show if the architecture extracts useful features and generalizes well.

Confusion matrix example

      Predicted
      | Cat | Dog |
    ---+-----+-----+
    Cat| 50  | 10  |
    Dog| 5   | 35  |

    TP (Cat) = 50, FP (Cat) = 10, FN (Cat) = 5, TN (Cat) = 35
    

This matrix helps calculate precision and recall for each class, showing how architecture impacts correct and wrong predictions.

Precision vs Recall tradeoff

A complex architecture might improve recall by finding more true objects but lower precision by adding false detections. A simpler design might have high precision but miss some objects (low recall). Choosing architecture depends on whether missing objects or false alarms are worse.

Example: In face recognition, high precision avoids false matches, but in medical image detection, high recall avoids missing diseases.

Good vs Bad metric values

Good: Accuracy above 90%, precision and recall balanced above 85%, F1 score high. This means the architecture captures features well and predicts reliably.

Bad: Accuracy high but recall very low (e.g., 40%), or precision very low. This shows the architecture misses many true cases or makes many false alarms, hurting performance.

Common pitfalls

• Overfitting: Complex architectures may memorize training data, showing high accuracy but poor real-world results.
• Data leakage: If test data leaks into training, metrics look falsely good, hiding architecture flaws.
• Ignoring class imbalance: Accuracy can be misleading if one class dominates; precision and recall give clearer insight.

Self-check question

Your model has 98% accuracy but only 12% recall on detecting a rare object. Is it good for production?

Answer: No. The model misses most true objects (low recall), so it fails its purpose despite high accuracy. The architecture likely does not capture important features for that object.

Key Result

Architecture design impacts key metrics like precision, recall, and F1 score, which reveal how well the model learns and generalizes.

Practice

1. Why does the design of a neural network architecture affect its performance on image tasks?

easy

A. Because it controls the size of the training dataset

B. Because it determines how well the model can learn important features from images

C. Because it decides the file format of the images

D. Because it changes the color of the images

Solution

1. Step 1: Understand the role of architecture in feature learning

   The architecture defines layers and connections that extract patterns from images.

2. Step 2: Connect architecture to model performance

   Better feature extraction leads to improved accuracy and generalization on tasks.

3. Final Answer:

   Because it determines how well the model can learn important features from images -> Option B

4. Quick Check:

   Architecture affects feature learning = D [OK]

Hint: Think about how model structure helps find image patterns [OK]

Common Mistakes:

• Confusing architecture with image properties
• Thinking architecture changes data format
• Believing architecture controls dataset size

2. Which of the following is the correct way to define a convolutional layer in a deep learning model using Python and PyTorch?

easy

A. nn.Conv2d(in_channels=3, out_channels=16, kernel_size=3, stride=1, padding=1)

B. nn.Linear(in_features=3, out_features=16)

C. nn.Conv1d(in_channels=3, out_channels=16, kernel_size=3)

D. nn.MaxPool2d(kernel_size=2, stride=2)

Solution

1. Step 1: Identify the convolutional layer syntax

   In PyTorch, Conv2d is used for 2D image convolutions with parameters for channels and kernel size.

2. Step 2: Check each option's layer type

   nn.Conv2d(in_channels=3, out_channels=16, kernel_size=3, stride=1, padding=1) correctly uses nn.Conv2d with proper parameters; others define different layers.

3. Final Answer:

   nn.Conv2d(in_channels=3, out_channels=16, kernel_size=3, stride=1, padding=1) -> Option A

4. Quick Check:

   Correct Conv2d syntax = B [OK]

Hint: Look for Conv2d with correct parameters for image layers [OK]

Common Mistakes:

• Confusing Conv2d with Linear or Conv1d layers
• Missing stride or padding parameters
• Choosing pooling layers instead of convolution

3. Consider this simplified CNN architecture for image classification:

model = nn.Sequential(
  nn.Conv2d(3, 8, 3, padding=1),
  nn.ReLU(),
  nn.MaxPool2d(2),
  nn.Conv2d(8, 16, 3, padding=1),
  nn.ReLU(),
  nn.MaxPool2d(2),
  nn.Flatten(),
  nn.Linear(16*8*8, 10)
)

If the input images are 32x32 pixels, what is the size of the feature map before flattening?

medium

A. 8 channels with 8x8 spatial size

B. 8 channels with 16x16 spatial size

C. 16 channels with 16x16 spatial size

D. 16 channels with 8x8 spatial size

Solution

1. Step 1: Calculate size after first Conv2d and MaxPool2d

   Input 32x32, Conv2d with padding=1 keeps size 32x32, MaxPool2d(2) halves to 16x16 with 8 channels.

2. Step 2: Calculate size after second Conv2d and MaxPool2d

   Conv2d keeps size 16x16 with 16 channels, MaxPool2d halves to 8x8 with 16 channels.

3. Final Answer:

   16 channels with 8x8 spatial size -> Option D

4. Quick Check:

   Pooling halves size twice = 8x8 with 16 channels [OK]

Hint: Each MaxPool2d(2) halves spatial size [OK]

Common Mistakes:

• Forgetting padding keeps size after convolution
• Not halving size after pooling
• Mixing channel counts with spatial dimensions

4. You have a CNN model that overfits training data but performs poorly on new images. Which architecture change can help reduce overfitting?

medium

A. Remove all pooling layers to keep more details

B. Increase the number of convolutional filters drastically

C. Add dropout layers to randomly ignore some neurons during training

D. Use a smaller batch size during training

Solution

1. Step 1: Understand overfitting and regularization

   Overfitting means the model memorizes training data; dropout helps by randomly ignoring neurons to generalize better.

2. Step 2: Evaluate options for reducing overfitting

   Adding dropout (A) is a common fix; increasing filters (B) may worsen overfitting; removing pooling (C) increases parameters; batch size (D) affects training stability but less impact on overfitting.

3. Final Answer:

   Add dropout layers to randomly ignore some neurons during training -> Option C

4. Quick Check:

   Dropout reduces overfitting = A [OK]

Hint: Use dropout to prevent memorizing training data [OK]

Common Mistakes:

• Thinking bigger models always reduce overfitting
• Removing pooling increases parameters and overfitting
• Confusing batch size effects with architecture changes

5. You want to design a model for real-time object detection on a mobile device. Which architectural choice best balances accuracy and speed?

hard

A. Use a lightweight architecture like MobileNet with depthwise separable convolutions

B. Use a very deep ResNet with 152 layers for highest accuracy

C. Use a fully connected network without convolutions

D. Use a large kernel size (e.g., 11x11) in all convolution layers

Solution

1. Step 1: Identify requirements for mobile real-time detection

   Mobile devices need fast, efficient models with good accuracy and low computation.

2. Step 2: Evaluate architectural options

   MobileNet uses depthwise separable convolutions to reduce computation while keeping accuracy; very deep ResNet is slow; fully connected networks lack spatial understanding; large kernels increase computation.

3. Final Answer:

   Use a lightweight architecture like MobileNet with depthwise separable convolutions -> Option A

4. Quick Check:

   MobileNet balances speed and accuracy = C [OK]

Hint: Choose lightweight models designed for mobile use [OK]

Common Mistakes:

• Picking very deep models ignoring speed constraints
• Using fully connected layers for images
• Choosing large kernels that slow down inference