14 million images, more than 20k categories, typically denoting prominent objects in the image, either common daily objects, or a wild range of animals. About 1 million of them also have bounding boxes for the objects.
Each image appears to have a single label associated to it. Care must have been taken somehow with categories, since some images contain severl possible objects, e.g. a person and some object.
In practice however, the ILSVRC subset is more commonly used.
Official project page: www.image-net.org/
The data license is restrictive and forbids commercial usage: www.image-net.org/download.php.
The categories are all part of WordNet, which means that there are several parent/child categories such as dog vs type of dog available. ImageNet1k only appears to have leaf nodes however (i.e. no "dog" label, just specific types of dog).
TODO vs COCO dataset.
As of v7:
- ~9M images
- 600 object classes
- bounding boxes
- visual relatoinships are really hard: storage.googleapis.com/openimages/web/factsfigures_v7.html#visual-relationships e.g. "person kicking ball": storage.googleapis.com/openimages/web/visualizer/index.html?type=relationships&set=train&c=kick
- google.github.io/localized-narratives/ localized narratives is ludicrous, you can actually hear the (Indian women mostly) annotators describing the image while hovering their mouses to point what they are talking about). They are clearly bored out of their minds the poor people!
OK, now we're talking, two liner and you get a window showing bounding box object detection from your webcam feed!The accuracy is crap for anything but people. But still. Well done. Tested on Ubuntu 22.10, P51.
python -m pip install -U yolov5==7.0.9
yolov5 detect --source 0