(a) Single objects:
The tasks used for single objects are Classification (assigning one label to the image) and Classification + Localisation (identifying what object is present and where it is located in the image).
(b) Multiple objects:
The tasks used for multiple objects are Object Detection (finding instances of real-world objects in images/videos) and Instance Segmentation (detecting object instances, assigning categories, and labelling each pixel).
Source: Chapter 5, Section 5.2 Computer Vision Tasks
---