Seeing the Unseen: Errors and Bias in Visual Datasets

Introduction From face recognition in smartphones to automatic routing on self-driving cars, machine vision algorithms lie in the core of these features. These systems solve image based tasks by identifying and understanding objects, subsequently making decisions from these information. A large set of images where the featured objects were labelled, known as datasets, are commonly used to develop and enhance machine vision algorithms (Cox 2016). However, errors in datasets are usually induced or even magnified in algorithms, at times resulting in issues such as recognising black people as gorillas and misrepresenting ethnicities in search results (Nieva 2015; Prabhu and Birhane 2020). This essay tracks the errors in datasets and their impacts, revealing that a flawed dataset could be a result of limited categories, incomprehensive sourcing and poor classification.