Lesson 1 · Computer Vision
What is Computer Vision?
6 min
You'll be able to
- Define computer vision and its goals
- Understand images as arrays of numbers
- Name the core vision tasks
Computer vision gives machines the ability to interpret the visual world. Because a camera image is just a grid of numbers — pixel intensities — we can use deep networks to extract meaning from them.
Core tasks
- Image classification — label the whole image.
- Object detection — find and locate objects.
- Segmentation — label each pixel.
- Face recognition — identify individuals.
import numpy as np
# 100x100 colour image: [100, 100, 3] uint8 values 0-255
image = np.random.randint(0, 255, (100, 100, 3), dtype=np.uint8)
print(image.shape) # (100, 100, 3)Challenge
See the grid
Describe how a greyscale photo of a landscape would look as an array of numbers — what do bright and dark regions map to?
Knowledge Check
Computer vision
A colour image with height 100, width 100 has the shape:
Image classification labels the whole image, while object detection also locates objects.
Answer all questions to submit.