์ฝ˜ํ…์ธ ๋กœ ๊ฑด๋„ˆ๋›ฐ๊ธฐ

๊ฐœ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ

์†Œ๊ฐœ

๊ฐœ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ๋Š” Ultralytics ๊ฐ•์•„์ง€ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ๋Š” ๊ฐ•์•„์ง€ ํ‚คํฌ์ธํŠธ ์ถ”์ •์„ ์œ„ํ•ด ํŠน๋ณ„ํžˆ ์„ ๋ณ„๋œ ๊ณ ํ’ˆ์งˆ์˜ ๊ด‘๋ฒ”์œ„ํ•œ ๋ฐ์ดํ„ฐ ์„ธํŠธ์ž…๋‹ˆ๋‹ค. 6,773๊ฐœ์˜ ํ›ˆ๋ จ ์ด๋ฏธ์ง€์™€ 1,703๊ฐœ์˜ ํ…Œ์ŠคํŠธ ์ด๋ฏธ์ง€๊ฐ€ ํฌํ•จ๋œ ์ด ๋ฐ์ดํ„ฐ ์„ธํŠธ๋Š” ๊ฐ•๋ ฅํ•œ ํฌ์ฆˆ ์ถ”์ • ๋ชจ๋ธ์„ ํ›ˆ๋ จํ•˜๊ธฐ ์œ„ํ•œ ํƒ„ํƒ„ํ•œ ํ† ๋Œ€๋ฅผ ์ œ๊ณตํ•ฉ๋‹ˆ๋‹ค. ๊ฐ ์ฃผ์„์ด ๋‹ฌ๋ฆฐ ์ด๋ฏธ์ง€์—๋Š” ํ‚คํฌ์ธํŠธ๋‹น 3์ฐจ์›(x, y, ๊ฐ€์‹œ์„ฑ)์˜ 24๊ฐœ ํ‚คํฌ์ธํŠธ๊ฐ€ ํฌํ•จ๋˜์–ด ์žˆ์–ด ์ปดํ“จํ„ฐ ๋น„์ „ ๋ถ„์•ผ์˜ ๊ณ ๊ธ‰ ์—ฐ๊ตฌ ๋ฐ ๊ฐœ๋ฐœ์— ์œ ์šฉํ•œ ๋ฆฌ์†Œ์Šค๋กœ ํ™œ์šฉํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.

Ultralytics ๊ฐœ ์ž์„ธ ๋””์Šคํ”Œ๋ ˆ์ด ์ด๋ฏธ์ง€

์ด ๋ฐ์ดํ„ฐ ์„ธํŠธ๋Š” Ultralytics HUB ๋ฐ YOLO11.

๋ฐ์ดํ„ฐ ์„ธํŠธ YAML

๋ฐ์ดํ„ฐ ์„ธํŠธ ๊ตฌ์„ฑ์„ ์ •์˜ํ•˜๋Š” ๋ฐ๋Š” YAML(๋˜ ๋‹ค๋ฅธ ๋งˆํฌ์—… ์–ธ์–ด) ํŒŒ์ผ์ด ์‚ฌ์šฉ๋ฉ๋‹ˆ๋‹ค. ์—ฌ๊ธฐ์—๋Š” ๊ฒฝ๋กœ, ํ‚คํฌ์ธํŠธ ์„ธ๋ถ€ ์ •๋ณด ๋ฐ ๊ธฐํƒ€ ๊ด€๋ จ ์ •๋ณด๊ฐ€ ํฌํ•จ๋ฉ๋‹ˆ๋‹ค. ๊ฐ•์•„์ง€ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ์˜ ๊ฒฝ์šฐ์—๋Š” dog-pose.yaml ์—์„œ ์‚ฌ์šฉํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. https://github.com/ultralytics/ultralytics/blob/main/ultralytics/cfg/datasets/dog-pose.yaml.

ultralytics/cfg/datasets/dog-pose.yaml

# Ultralytics YOLO ๐Ÿš€, AGPL-3.0 license
# Dogs dataset http://vision.stanford.edu/aditya86/ImageNetDogs/ by Stanford
# Documentation: https://docs.ultralytics.com/datasets/pose/dog-pose/
# Example usage: yolo train data=dog-pose.yaml
# parent
# โ”œโ”€โ”€ ultralytics
# โ””โ”€โ”€ datasets
#     โ””โ”€โ”€ dog-pose  โ† downloads here (337 MB)

# Train/val/test sets as 1) dir: path/to/imgs, 2) file: path/to/imgs.txt, or 3) list: [path/to/imgs1, path/to/imgs2, ..]
path: ../datasets/dog-pose # dataset root dir
train: train # train images (relative to 'path') 6773 images
val: val # val images (relative to 'path') 1703 images

# Keypoints
kpt_shape: [24, 3] # number of keypoints, number of dims (2 for x,y or 3 for x,y,visible)

# Classes
names:
  0: dog

# Download script/URL (optional)
download: https://github.com/ultralytics/assets/releases/download/v0.0.0/dog-pose.zip

์‚ฌ์šฉ๋ฒ•

์ด๋ฏธ์ง€ ํฌ๊ธฐ๊ฐ€ 640์ธ 100๊ฐœ์˜ ๊ฐœ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ์— ๋Œ€ํ•ด YOLO11n ํฌ์ฆˆ ๋ชจ๋ธ์„ ํ›ˆ๋ จํ•˜๋ ค๋ฉด ๋‹ค์Œ ์ฝ”๋“œ ์กฐ๊ฐ์„ ์‚ฌ์šฉํ•  ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์‚ฌ์šฉ ๊ฐ€๋Šฅํ•œ ์ธ์ˆ˜์˜ ์ „์ฒด ๋ชฉ๋ก์€ ๋ชจ๋ธ ํ›ˆ๋ จ ํŽ˜์ด์ง€๋ฅผ ์ฐธ์กฐํ•˜์„ธ์š”.

์—ด์ฐจ ์˜ˆ์‹œ

from ultralytics import YOLO

# Load a model
model = YOLO("yolo11n-pose.pt")  # load a pretrained model (recommended for training)

# Train the model
results = model.train(data="dog-pose.yaml", epochs=100, imgsz=640)
# Start training from a pretrained *.pt model
yolo pose train data=dog-pose.yaml model=yolo11n-pose.pt epochs=100 imgsz=640

์ƒ˜ํ”Œ ์ด๋ฏธ์ง€ ๋ฐ ์ฃผ์„

๋‹ค์Œ์€ ๊ฐœ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ์˜ ์ด๋ฏธ์ง€ ๋ช‡ ๊ฐ€์ง€ ์˜ˆ์™€ ํ•ด๋‹น ์ฃผ์„์ž…๋‹ˆ๋‹ค:

๋ฐ์ดํ„ฐ ์„ธํŠธ ์ƒ˜ํ”Œ ์ด๋ฏธ์ง€

  • ๋ชจ์ž์ดํฌ ์ด๋ฏธ์ง€: ์ด ์ด๋ฏธ์ง€๋Š” ๋ชจ์ž์ดํฌ๋œ ๋ฐ์ดํ„ฐ ์„ธํŠธ ์ด๋ฏธ์ง€๋กœ ๊ตฌ์„ฑ๋œ ํ›ˆ๋ จ ๋ฐฐ์น˜์˜ ์˜ˆ์‹œ์ž…๋‹ˆ๋‹ค. ๋ชจ์ž์ดํฌ๋Š” ์—ฌ๋Ÿฌ ์ด๋ฏธ์ง€๋ฅผ ํ•˜๋‚˜์˜ ์ด๋ฏธ์ง€๋กœ ๊ฒฐํ•ฉํ•˜์—ฌ ๊ฐ ํ›ˆ๋ จ ๋ฐฐ์น˜ ๋‚ด์—์„œ ๋‹ค์–‘ํ•œ ๊ฐœ์ฒด์™€ ์žฅ๋ฉด์„ ๋Š˜๋ฆฌ๊ธฐ ์œ„ํ•ด ํ›ˆ๋ จ ์ค‘์— ์‚ฌ์šฉ๋˜๋Š” ๊ธฐ์ˆ ์ž…๋‹ˆ๋‹ค. ์ด๋ฅผ ํ†ตํ•ด ๋‹ค์–‘ํ•œ ๊ฐ์ฒด ํฌ๊ธฐ, ์ข…ํšก๋น„ ๋ฐ ์ปจํ…์ŠคํŠธ์— ์ผ๋ฐ˜ํ™”ํ•˜๋Š” ๋ชจ๋ธ์˜ ๋Šฅ๋ ฅ์„ ํ–ฅ์ƒ์‹œํ‚ฌ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค.

์ด ์˜ˆ๋Š” ๊ฐœ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ์— ํฌํ•จ๋œ ์ด๋ฏธ์ง€์˜ ๋‹ค์–‘์„ฑ๊ณผ ๋ณต์žก์„ฑ, ๊ทธ๋ฆฌ๊ณ  ํ›ˆ๋ จ ๊ณผ์ •์—์„œ ๋ชจ์ž์ดํ‚น์„ ์‚ฌ์šฉํ•  ๋•Œ์˜ ์ด์ ์„ ๋ณด์—ฌ์ค๋‹ˆ๋‹ค.

์ธ์šฉ ๋ฐ ๊ฐ์‚ฌ

์—ฐ๊ตฌ ๋˜๋Š” ๊ฐœ๋ฐœ ์ž‘์—…์— ๊ฐœ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ๋ฅผ ์‚ฌ์šฉํ•˜๋Š” ๊ฒฝ์šฐ ๋‹ค์Œ ๋…ผ๋ฌธ์„ ์ธ์šฉํ•ด ์ฃผ์„ธ์š”:

@inproceedings{khosla2011fgvc,
  title={Novel dataset for Fine-Grained Image Categorization},
  author={Aditya Khosla and Nityananda Jayadevaprakash and Bangpeng Yao and Li Fei-Fei},
  booktitle={First Workshop on Fine-Grained Visual Categorization (FGVC), IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2011}
}
@inproceedings{deng2009imagenet,
  title={ImageNet: A Large-Scale Hierarchical Image Database},
  author={Jia Deng and Wei Dong and Richard Socher and Li-Jia Li and Kai Li and Li Fei-Fei},
  booktitle={IEEE Computer Vision and Pattern Recognition (CVPR)},
  year={2009}
}

์ปดํ“จํ„ฐ ๋น„์ „ ์ปค๋ฎค๋‹ˆํ‹ฐ๋ฅผ ์œ„ํ•ด ์ด ๊ท€์ค‘ํ•œ ๋ฆฌ์†Œ์Šค๋ฅผ ๋งŒ๋“ค๊ณ  ์œ ์ง€ ๊ด€๋ฆฌํ•ด ์ฃผ์‹  ์Šคํƒ ํฌ๋“œ ํŒ€์— ๊ฐ์‚ฌ์˜ ๋ง์”€์„ ์ „ํ•ฉ๋‹ˆ๋‹ค. ๊ฐœ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ์™€ ์ œ์ž‘์ž์— ๋Œ€ํ•œ ์ž์„ธํ•œ ๋‚ด์šฉ์€ ์Šคํƒ ํฌ๋“œ ๊ฐœ ๋ฐ์ดํ„ฐ ์„ธํŠธ ์›น์‚ฌ์ดํŠธ๋ฅผ ๋ฐฉ๋ฌธํ•˜์„ธ์š”.

์ž์ฃผ ๋ฌป๋Š” ์งˆ๋ฌธ

๋„๊ทธ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ๋Š” ๋ฌด์—‡์ด๋ฉฐ Ultralytics YOLO11 ?

๊ฐœ ์ž์„ธ ๋ฐ์ดํ„ฐ ์„ธํŠธ์—๋Š” ๊ฐœ ์ž์„ธ ์ถ”์ •์„ ์œ„ํ•œ 17๊ฐœ์˜ ํ‚คํฌ์ธํŠธ๋กœ ์ฃผ์„์ด ๋‹ฌ๋ฆฐ 6,000๊ฐœ์˜ ์ด๋ฏธ์ง€๊ฐ€ ํฌํ•จ๋˜์–ด ์žˆ์Šต๋‹ˆ๋‹ค. ๋‹ค์Œ์„ ์‚ฌ์šฉํ•˜์—ฌ ๋ชจ๋ธ์„ ํ›ˆ๋ จํ•˜๊ณ  ๊ฒ€์ฆํ•˜๋Š” ๋ฐ ์ด์ƒ์ ์ž…๋‹ˆ๋‹ค. Ultralytics YOLO11๋ฅผ ์‚ฌ์šฉํ•œ ๋ชจ๋ธ ํ›ˆ๋ จ ๋ฐ ๊ฒ€์ฆ์— ์ด์ƒ์ ์ž…๋‹ˆ๋‹ค. ๋™๋ฌผ ํ–‰๋™ ๋ถ„์„ ๋ฐ ์ˆ˜์˜ํ•™ ์—ฐ๊ตฌ์™€ ๊ฐ™์€ ์• ํ”Œ๋ฆฌ์ผ€์ด์…˜์„ ์ง€์›ํ•ฉ๋‹ˆ๋‹ค.

Ultralytics ์˜ ๊ฐœ ์ž์„ธ ๋ฐ์ดํ„ฐ ์ง‘ํ•ฉ์„ ์‚ฌ์šฉํ•˜์—ฌ YOLO11 ๋ชจ๋ธ์„ ํ›ˆ๋ จํ•˜๋ ค๋ฉด ์–ด๋–ป๊ฒŒ ํ•˜๋‚˜์š”?

์ด๋ฏธ์ง€ ํฌ๊ธฐ๊ฐ€ 640์ธ 100๊ฐœ์˜ ๊ฐœ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ์— ๋Œ€ํ•ด YOLO11n-ํฌ์ฆˆ ๋ชจ๋ธ์„ ํ›ˆ๋ จํ•˜๋ ค๋ฉด ๋‹ค์Œ ์˜ˆ์ œ๋ฅผ ๋”ฐ๋ฅด์„ธ์š”:

์—ด์ฐจ ์˜ˆ์‹œ

from ultralytics import YOLO

# Load a model
model = YOLO("yolo11n-pose.pt")

# Train the model
results = model.train(data="dog-pose.yaml", epochs=100, imgsz=640)
yolo pose train data=dog-pose.yaml model=yolo11n-pose.pt epochs=100 imgsz=640

ํ›ˆ๋ จ ์ธ์ˆ˜์˜ ์ „์ฒด ๋ชฉ๋ก์€ ๋ชจ๋ธ ํ›ˆ๋ จ ํŽ˜์ด์ง€๋ฅผ ์ฐธ์กฐํ•˜์„ธ์š”.

๊ฐ•์•„์ง€ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ๋ฅผ ์‚ฌ์šฉํ•˜๋ฉด ์–ด๋–ค ์ด์ ์ด ์žˆ๋‚˜์š”?

๊ฐœ ํฌ์ฆˆ ๋ฐ์ดํ„ฐ ์„ธํŠธ๋Š” ๋ช‡ ๊ฐ€์ง€ ์ด์ ์„ ์ œ๊ณตํ•ฉ๋‹ˆ๋‹ค:

๋ฐฉ๋Œ€ํ•˜๊ณ  ๋‹ค์–‘ํ•œ ๋ฐ์ดํ„ฐ ์„ธํŠธ: 6,000๊ฐœ์˜ ์ด๋ฏธ์ง€๋กœ ๋‹ค์–‘ํ•œ ๋ฐ˜๋ ค๊ฒฌ์˜ ์ž์„ธ, ํ’ˆ์ข…, ์ƒํ™ฉ์„ ์•„์šฐ๋ฅด๋Š” ๋ฐฉ๋Œ€ํ•œ ์–‘์˜ ๋ฐ์ดํ„ฐ๋ฅผ ์ œ๊ณตํ•˜์—ฌ ๊ฐ•๋ ฅํ•œ ๋ชจ๋ธ ํ›ˆ๋ จ ๋ฐ ํ‰๊ฐ€๊ฐ€ ๊ฐ€๋Šฅํ•ฉ๋‹ˆ๋‹ค.

ํฌ์ฆˆ๋ณ„ ์ฃผ์„: ํฌ์ฆˆ ์ถ”์ •์„ ์œ„ํ•œ ์ƒ์„ธํ•œ ์ฃผ์„์„ ์ œ๊ณตํ•˜์—ฌ ํฌ์ฆˆ ๊ฐ์ง€ ๋ชจ๋ธ ํ•™์Šต์„ ์œ„ํ•œ ๊ณ ํ’ˆ์งˆ ๋ฐ์ดํ„ฐ๋ฅผ ๋ณด์žฅํ•ฉ๋‹ˆ๋‹ค.

์‹ค์ œ ์‹œ๋‚˜๋ฆฌ์˜ค: ๋‹ค์–‘ํ•œ ํ™˜๊ฒฝ์˜ ์ด๋ฏธ์ง€๋ฅผ ํฌํ•จํ•˜์—ฌ ์‹ค์ œ ์• ํ”Œ๋ฆฌ์ผ€์ด์…˜์— ์ผ๋ฐ˜ํ™”ํ•  ์ˆ˜ ์žˆ๋Š” ๋ชจ๋ธ์˜ ๊ธฐ๋Šฅ์„ ํ–ฅ์ƒ์‹œํ‚ต๋‹ˆ๋‹ค.

๋ชจ๋ธ ์„ฑ๋Šฅ ํ–ฅ์ƒ: ๋ฐ์ดํ„ฐ ์„ธํŠธ์˜ ๋‹ค์–‘์„ฑ๊ณผ ๊ทœ๋ชจ๋Š” ํŠนํžˆ ์„ธ๋ฐ€ํ•œ ํฌ์ฆˆ ์ถ”์ •๊ณผ ๊ด€๋ จ๋œ ์ž‘์—…์—์„œ ๋ชจ๋ธ ์ •ํ™•๋„์™€ ๊ฒฌ๊ณ ์„ฑ์„ ๊ฐœ์„ ํ•˜๋Š” ๋ฐ ๋„์›€์ด ๋ฉ๋‹ˆ๋‹ค.

๊ธฐ๋Šฅ ๋ฐ ์‚ฌ์šฉ๋ฒ•์— ๋Œ€ํ•œ ์ž์„ธํ•œ ๋‚ด์šฉ์€ ๋ฐ์ดํ„ฐ ์„ธํŠธ ์†Œ๊ฐœ ์„น์…˜์„ ์ฐธ์กฐํ•˜์„ธ์š”.

๋ชจ์ž์ดํฌ๊ฐ€ ๊ฐœ ์ž์„ธ ๋ฐ์ดํ„ฐ ์„ธํŠธ๋ฅผ ์‚ฌ์šฉํ•˜๋Š” YOLO11 ํ›ˆ๋ จ ํ”„๋กœ์„ธ์Šค์— ์–ด๋–ค ์ด์ ์ด ์žˆ๋‚˜์š”?

๊ฐœ ์ž์„ธ ๋ฐ์ดํ„ฐ ์„ธํŠธ์˜ ์ƒ˜ํ”Œ ์ด๋ฏธ์ง€์—์„œ ๋ณผ ์ˆ˜ ์žˆ๋“ฏ์ด ๋ชจ์ž์ด์‹ฑ์€ ์—ฌ๋Ÿฌ ์ด๋ฏธ์ง€๋ฅผ ํ•˜๋‚˜์˜ ํ•ฉ์„ฑ ์ด๋ฏธ์ง€๋กœ ๋ณ‘ํ•ฉํ•˜์—ฌ ๊ฐ ํ›ˆ๋ จ ๋ฐฐ์น˜์—์„œ ๋ฌผ์ฒด์™€ ์žฅ๋ฉด์˜ ๋‹ค์–‘์„ฑ์„ ํ’๋ถ€ํ•˜๊ฒŒ ํ•ฉ๋‹ˆ๋‹ค. ์ด ์ ‘๊ทผ ๋ฐฉ์‹์€ ๋‹ค์–‘ํ•œ ๊ฐ์ฒด ํฌ๊ธฐ, ์ข…ํšก๋น„ ๋ฐ ์ปจํ…์ŠคํŠธ์— ๊ฑธ์ณ ์ผ๋ฐ˜ํ™”ํ•˜๋Š” ๋ชจ๋ธ์˜ ๋Šฅ๋ ฅ์„ ํ–ฅ์ƒ์‹œ์ผœ ์„ฑ๋Šฅ์„ ๊ฐœ์„ ํ•ฉ๋‹ˆ๋‹ค. ์ด๋ฏธ์ง€ ์˜ˆ์‹œ๋Š” ์ƒ˜ํ”Œ ์ด๋ฏธ์ง€ ๋ฐ ์ฃผ์„ ์„น์…˜์„ ์ฐธ์กฐํ•˜์„ธ์š”.

Dog-pose ๋ฐ์ดํ„ฐ ์„ธํŠธ YAML ํŒŒ์ผ์€ ์–ด๋””์—์„œ ์ฐพ์„ ์ˆ˜ ์žˆ์œผ๋ฉฐ ์–ด๋–ป๊ฒŒ ์‚ฌ์šฉํ•˜๋‚˜์š”?

Dog-pose ๋ฐ์ดํ„ฐ์„ธํŠธ YAML ํŒŒ์ผ์€ ์—ฌ๊ธฐ์—์„œ ์ฐพ์„ ์ˆ˜ ์žˆ์Šต๋‹ˆ๋‹ค. ์ด ํŒŒ์ผ์€ ๊ฒฝ๋กœ, ํด๋ž˜์Šค ๋ฐ ๊ธฐํƒ€ ๊ด€๋ จ ์ •๋ณด๋ฅผ ํฌํ•จํ•œ ๋ฐ์ดํ„ฐ ์„ธํŠธ ๊ตฌ์„ฑ์„ ์ •์˜ํ•ฉ๋‹ˆ๋‹ค. ์ด ํŒŒ์ผ์„ ํ›ˆ๋ จ ์˜ˆ์ œ ์„น์…˜์— ์–ธ๊ธ‰๋œ YOLO11 ํ›ˆ๋ จ ์Šคํฌ๋ฆฝํŠธ์™€ ํ•จ๊ป˜ ์‚ฌ์šฉํ•˜์„ธ์š”.

๋” ๋งŽ์€ FAQ์™€ ์ž์„ธํ•œ ์„ค๋ช…์„œ๋ฅผ ๋ณด๋ ค๋ฉด Ultralytics ๋ฌธ์„œ๋ฅผ ์ฐธ์กฐํ•˜์„ธ์š”.

๐Ÿ“… Created 23 days ago โœ๏ธ Updated 23 days ago

๋Œ“๊ธ€