The Microsoft COCO dataset is often used to benchmark and consider pc imaginative and prescient mannequin architectures. It’s also generally used to coach “base weights” which you could fine-tune utilizing customized knowledge utilizing switch studying. The newest COCO problem in 2020 included knowledge for object detection, keypoint detection, panoptic segmentation, and dense pose estimation.
You may strive a YOLO11 mannequin skilled on Microsoft COCO within the following playground. Drag in a picture that accommodates an object that Microsoft COCO can determine to see how the mannequin performs:
For object detection, Microsoft COCO can reference 80 lessons. These are, in alphabetical order:
airplane
apple
backpack
banana
baseball hat
baseball glove
bear
mattress
bench
bicycle
hen
boat
e-book
bottle
bowl
broccoli
bus
cake
automotive
carrot
cat
mobile phone
chair clock
sofa
cow
cup
eating desk
canine
donut
elephant
hearth hydrant
fork
frisbee
giraffe
hair drier
purse
horse
sizzling canine
keyboard
kite
knife
laptop computer
microwave
motorbike
mouse
orange
oven
parking meter
particular person
pizza
potted plant
fridge
distant
sandwich
scissors
sheep
sink
skateboard
skis
snowboard
spoon
sports activities ball
cease signal
suitcase
surfboard
teddy bear
tennis racket
tie
toaster
bathroom
toothbrush
visitors mild
prepare
truch
television
umbrella
vase
wine glassTo study extra about Microsoft COCO, seek advice from our Introduction to the COCO dataset
You can too strive fashions skilled on the Microsoft COCO dataset utilizing varied architectures on the Roboflow Univese Microsoft COCO playground.
Cite this Put up
Use the next entry to quote this put up in your analysis:
James Gallagher. (Oct 15, 2024). Microsoft COCO Lessons Reference Checklist. Roboflow Weblog: https://weblog.roboflow.com/microsoft-coco-classes/
