Frame extraction in Computer Vision - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of frame extraction loop

What will be the output of the following Python code snippet that extracts frames from a video using OpenCV?

Computer Vision

import cv2
cap = cv2.VideoCapture('video.mp4')
count = 0
while True:
    ret, frame = cap.read()
    if not ret or count == 3:
        break
    print(f'Frame {count} shape:', frame.shape)
    count += 1
cap.release()

A

Frame 0 shape: (640, 480, 3)
Frame 1 shape: (640, 480, 3)
Frame 2 shape: (640, 480, 3)

B

Frame 0 shape: (480, 640, 3)
Frame 1 shape: (480, 640, 3)
Frame 2 shape: (480, 640, 3)

C

Frame 0 shape: (480, 640)
Frame 1 shape: (480, 640)
Frame 2 shape: (480, 640)

D No output, the loop breaks immediately

💡 Hint

Remember that OpenCV loads color images with shape (height, width, channels).

OpenCV reads frames as color images with shape (height, width, 3). The loop prints the shape for the first 3 frames.

🧠 Conceptual

intermediate

Best frame extraction method for uniform sampling

You want to extract exactly 10 frames evenly spaced from a 100-frame video. Which method is best?

A Extract frames randomly from the video

B Extract the first 10 frames only

C Extract frames at indices 0, 10, 20, ..., 90

D Extract every frame and then select 10 randomly

💡 Hint

Think about how to get frames evenly spaced across the whole video.

Extracting frames at fixed intervals (every 10 frames) ensures uniform coverage of the video.

❓ Hyperparameter

advanced

Choosing frame extraction rate for motion analysis

For analyzing fast motion in a video, which frame extraction rate is most suitable?

A Extract 1 frame per second

B Extract 1 frame every 5 seconds

C Extract frames randomly

D Extract every frame (full frame rate)

💡 Hint

Fast motion requires detailed temporal information.

Extracting every frame preserves all temporal details needed for fast motion analysis.

🔧 Debug

advanced

Identify error in frame extraction code

What error will this code raise when extracting frames from a video?

Computer Vision

import cv2
cap = cv2.VideoCapture('video.mp4')
frame = cap.read()[1]
print(frame.shape)
cap.release()

A AttributeError: 'NoneType' object has no attribute 'shape'

B SyntaxError: invalid syntax

C TypeError: 'tuple' object is not callable

D No error, prints frame shape

💡 Hint

Check what cap.read() returns and if the frame is valid.

cap.read() returns (ret, frame). If ret is False, frame is None, so accessing shape causes AttributeError.

❓ Model Choice

expert

Best model for frame-based action recognition

You want to classify actions in video clips by analyzing extracted frames. Which model architecture is best suited?

A 3D CNN that processes multiple frames together

B K-means clustering on frame histograms

C Fully connected network on raw pixel values

D 2D CNN applied independently on each frame

💡 Hint

Consider models that capture spatial and temporal information.

3D CNNs process spatial and temporal dimensions jointly, making them ideal for action recognition from frames.