They specify a list of points of interest, connections between those points, where those points are within the segmentation, and whether the points are visible.

Keypoints add additional information about a segmented object. Is a category id of 1 (which is a person)Ĭorresponds with an image with id 250282 (which is a vintage class photo of about 50 school children) Is a crowd (meaning it’s a group of objects) Has an area of 220834 pixels (much larger) and a bounding box of Has a Run-Length-Encoding style segmentation Is not a crowd (meaning it’s a single object)Ĭorresponds with an image with id 289343 (which is a person on a strange bicycle and a tiny dog) Has an area of 702 pixels (pretty small) and a bounding box of Has a segmentation list of vertices (x, y pixel positions) The following JSON shows 2 different annotations. The category id corresponds to a single category specified in the categories section.Įach annotation also has an id (unique to all other annotations in the dataset). The image id corresponds to a specific image in the dataset. Is Crowd specifies whether the segmentation is for a single object or for a group/cluster of objects. a 10px by 20px box would have an area of 200). I’ll explain how this works later in the article.Īrea is measured in pixels (e.g. Typically, RLE is used for groups of objects (like a large stack of books). I say “usually” because regions of interest indicated by these annotations are specified by “segmentations”, which are usually a list of polygon vertices around the object, but can also be a run-length-encoded (RLE) bit mask. Usually this results in a new annotation item for each one. Often there will be multiple instances of an object in an image. For example, if there are 64 bicycles spread out across 100 images, there will be 64 bicycle annotations (along with a ton of annotations for other object categories). It contains a list of every individual object annotation from every image in the dataset. The “annotations” section is the trickiest to understand. The COCO dataset is formatted in JSON and is a collection of “info”, “licenses”, “images”, “annotations”, “categories” (in most cases), and “segment info” (in one case). If you’re not a video person or want more detail, keep reading.

Coc 3d add