Understanding the dataset
In this section, we will cover the dataset on which the Deep Learning models have been trained. There are two datasets that are heavily used when we are trying to build the object detection application, and those datasets are as follows:
The COCO dataset
The PASCAL VOC dataset
We will look at each of the datasets one by one.
The COCO dataset
COCO stands for Common object in context. So, the short form for this dataset is the COCO dataset. Many tech giants, such as Google, Facebook, Microsoft, and so on are using COCO data to build amazing applications for object detection, object segmentation, and so on. You can find details regarding this dataset at this official web page:
The COCO dataset is a large-scale object detection, segmentation, and captioning dataset. In this dataset, there are a total of 330,000 images, of which more than 200,000 are labeled. These images contain 1.5 million object instances with 80 object categories. All the...
