Object detection

As explained in Math and Architectures of Deep Learning

A computer-vision task of identifying object categories in an image and estimating their spatial locations, often with bounding boxes.

Object detection identifies the categories of objects in an image and estimates where they appear. Unlike image classification, which assigns categories to an image, detection also localizes each object with a bounding box. section 11.0

A bounding box represents an object's location with four coordinates:

• [(xₜₗ, yₜₗ), (xᵦᵣ, yᵦᵣ)], where the first point is the top-left corner and the second is the bottom-right corner. section 11.0
• The box sides remain parallel to the image coordinate axes. section 11.0

A detector must handle variation in object size, shape, color, viewpoint, lighting, and background. An early deep-learning strategy, R-CNN, extracted candidate regions, classified each region, and selected regions with high confidence. section 11.0 section 11.3

From recognition to detection

An image containing a bird, a plane, and Superman illustrates category recognition. Detection additionally reports each object's bounding-box coordinates; an image with four animals requires repeating this classification-and-localization process for every animal. section 11.0 section 11.3

Figure 11.1 Is it a bird? Is it a plane? Is it Superman?

Use the figure to distinguish identifying visible object categories from locating those objects within the image. section 11.0

How Math and Architectures of Deep Learning covers it

How deep it goes

Defines bounding-box coordinates and sketches R-CNN region classification, without detailing detector training or later architectures.

Examples

Uses images containing a bird, plane, Superman, and four stacked animals to illustrate recognition and localization.

What it assumes

Assumes image classification as the starting point and teaches object localization through bounding-box coordinates.

What you can do after

Evaluate/understand how detectors extend image classification by assigning categories and bounding boxes to objects.

How it connects

Connects object detection with image classification, bounding boxes, convolutional neural networks, and R-CNN.

liveBookmanning.com

Read Math and Architectures of Deep Learning on liveBook Get the book at manning.com

Also explained in

• 

  liveBookmanning.com
  Machine Learning Platform EngineeringObject detection is a computer-vision task that identifies objects in an image or other visual input and locates them with bounding regions.
• 

  liveBookmanning.com
  Inside Deep Learning: Math, Algorithms, ModelsA vision task in which a system identifies and localizes multiple objects in a single image, often producing several predictions per image.
• 

  liveBookmanning.com
  Hugging Face in ActionObject detection is the machine learning task of identifying and locating objects within images or video.
• 

  liveBookmanning.com
  Designing Deep Learning Systems: A guide for software engineersObject detection is the computer-vision task of locating instances of target object classes in images and assigning labels to them.

Compare how 5 books explain object detection →

Related concepts

• image segmentationcontrasts with · Image segmentation is a computer-vision task that partitions an image and assigns labels to its pixels or regions to delineate objects or meaningful areas.
• computer visionpart of · Computer vision is the field of computing concerned with using computational methods to extract and interpret information from images, video, and other visual data.
• bounding boxuses · A bounding box is an axis-aligned rectangular or hyperrectangular region defined by coordinate limits that encloses or localizes an object, area, or set of points.
• confidencemeasured by · Confidence is a degree of certainty or assurance associated with a judgment, estimate, prediction, capability, or statistical procedure, often expressed as a score, probability, or coverage level.

This explainer was generated by Manning's knowledge graph from the text of Math and Architectures of Deep Learning. The section links point to the parts of the book it draws on.