Machine learning evolving computer vision: Estimating socioeconomic levels using satellite imagery
Mario Abarca, Moisés Arizpe, Horacio González Duhart, Gustavo Ramírez and Marco Iván Zepeda
Kantar Mexico, a market researcher, explains how, by using widely available datasets, a state-of-the-art neural model and business knowledge, high value can be extracted at low cost from visual data.
Preface
How to extract great insights from satellite imagery? The last few years have seen enormous growth on Al-based data-enrichment techniques. Automatic sentiment and topic identification in text documents have become ubiquitous. Image classificat