I am a bit confused by reading A survey on object detection in remote sensing. They state that machine learning-based object detection consists of three essential parts - feature extraction, feature fusion + dimension reduction, and classifier training. Then, they list the feature extraction methods:
- Histogram of Oriented Gradients,
- Bag of Words,
- Texture Features,
- and more.
Later in the section, they list approaches to the classifier training, e.g.:
- neural networks.
This does not align with my understanding of things. My experience is mostly with CNNs. When I use convolutional neural networks for object detection, they consist of a feature extractor and a classifier, don't they? And the feature extractor are the convolutional (and/or other) layers. I've never come across an algorithm/technique that uses e.g. Bag of Words to extract features and then feeds this to a neural net. I can imagine e.g. Bag of Words and k-nearest-neighbors or SVM. But I thought that deep neural nets' main advantage is that we don't have to do the prior feature engineering.
Question: Does the article (the part I mention here) accurately describe the state of object detection machine learning models? Is the article just outdated (it's from 2016)? Or am I missing something