coco-viewer

coco-viewer copied to clipboard

hi @trsvchn, does segmentation element in JSON file needs to be written in RLE format, polygon or it can be just pixel coordinates of my ground truth?

@ivand9, thank you for your question!

coco-viewer currently works only with coco format for "Object Detection" task with the following structure

Each object instance annotation contains a series of fields, including the category id and segmentation mask of the object. The segmentation format depends on whether the instance represents a single object (iscrowd=0 in which case polygons are used) or a collection of objects (iscrowd=1 in which case RLE is used). Note that a single object (iscrowd=0) may require multiple polygons, for example if occluded.

annotation {
  "id" : int,
  "image_id" : int,
  "category_id" : int,
  "segmentation" : RLE or [polygon],
  "area" : float,
  "bbox" : [x,y,width,height],
"iscrowd" : 0 or 1,
}

more on coco data format here Data format

so it's RLE or polygon or list of polygons