μ½˜ν…μΈ λ‘œ κ±΄λ„ˆλ›°κΈ°

COCO8-Seg 데이터 μ„ΈνŠΈ

μ†Œκ°œ

Ultralytics COCO8-Seg is a small, but versatile instance segmentation dataset composed of the first 8 images of the COCO train 2017 set, 4 for training and 4 for validation. This dataset is ideal for testing and debugging segmentation models, or for experimenting with new detection approaches. With 8 images, it is small enough to be easily manageable, yet diverse enough to test training pipelines for errors and act as a sanity check before training larger datasets.

This dataset is intended for use with Ultralytics HUB and YOLO11.

데이터 μ„ΈνŠΈ YAML

데이터 μ„ΈνŠΈ ꡬ성을 μ •μ˜ν•˜λŠ” λ°λŠ” YAML(또 λ‹€λ₯Έ λ§ˆν¬μ—… μ–Έμ–΄) 파일이 μ‚¬μš©λ©λ‹ˆλ‹€. μ—¬κΈ°μ—λŠ” 데이터 μ„ΈνŠΈμ˜ 경둜, 클래슀 및 기타 κ΄€λ ¨ 정보에 λŒ€ν•œ 정보가 ν¬ν•¨λ˜μ–΄ μžˆμŠ΅λ‹ˆλ‹€. COCO8-Seg 데이터 μ„ΈνŠΈμ˜ 경우, 데이터 μ„ΈνŠΈμ˜ coco8-seg.yaml νŒŒμΌμ€ λ‹€μŒ μœ„μΉ˜μ—μ„œ μœ μ§€λ©λ‹ˆλ‹€. https://github.com/ultralytics/ultralytics/blob/main/ultralytics/cfg/datasets/coco8-seg.yaml.

ultralytics/cfg/datasets/coco8-seg.yaml

# Ultralytics YOLO πŸš€, AGPL-3.0 license
# COCO8-seg dataset (first 8 images from COCO train2017) by Ultralytics
# Documentation: https://docs.ultralytics.com/datasets/segment/coco8-seg/
# Example usage: yolo train data=coco8-seg.yaml
# parent
# β”œβ”€β”€ ultralytics
# └── datasets
#     └── coco8-seg  ← downloads here (1 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/coco8-seg # dataset root dir
train: images/train # train images (relative to 'path') 4 images
val: images/val # val images (relative to 'path') 4 images
test: # test images (optional)

# Classes
names:
  0: person
  1: bicycle
  2: car
  3: motorcycle
  4: airplane
  5: bus
  6: train
  7: truck
  8: boat
  9: traffic light
  10: fire hydrant
  11: stop sign
  12: parking meter
  13: bench
  14: bird
  15: cat
  16: dog
  17: horse
  18: sheep
  19: cow
  20: elephant
  21: bear
  22: zebra
  23: giraffe
  24: backpack
  25: umbrella
  26: handbag
  27: tie
  28: suitcase
  29: frisbee
  30: skis
  31: snowboard
  32: sports ball
  33: kite
  34: baseball bat
  35: baseball glove
  36: skateboard
  37: surfboard
  38: tennis racket
  39: bottle
  40: wine glass
  41: cup
  42: fork
  43: knife
  44: spoon
  45: bowl
  46: banana
  47: apple
  48: sandwich
  49: orange
  50: broccoli
  51: carrot
  52: hot dog
  53: pizza
  54: donut
  55: cake
  56: chair
  57: couch
  58: potted plant
  59: bed
  60: dining table
  61: toilet
  62: tv
  63: laptop
  64: mouse
  65: remote
  66: keyboard
  67: cell phone
  68: microwave
  69: oven
  70: toaster
  71: sink
  72: refrigerator
  73: book
  74: clock
  75: vase
  76: scissors
  77: teddy bear
  78: hair drier
  79: toothbrush

# Download script/URL (optional)
download: https://github.com/ultralytics/assets/releases/download/v0.0.0/coco8-seg.zip

μ‚¬μš©λ²•

To train a YOLO11n-seg model on the COCO8-Seg dataset for 100 epochs with an image size of 640, you can use the following code snippets. For a comprehensive list of available arguments, refer to the model Training page.

μ—΄μ°¨ μ˜ˆμ‹œ

from ultralytics import YOLO

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

# Train the model
results = model.train(data="coco8-seg.yaml", epochs=100, imgsz=640)
# Start training from a pretrained *.pt model
yolo segment train data=coco8-seg.yaml model=yolo11n-seg.pt epochs=100 imgsz=640

μƒ˜ν”Œ 이미지 및 주석

λ‹€μŒμ€ COCO8-Seg 데이터 μ„ΈνŠΈμ˜ 이미지와 ν•΄λ‹Ή μ£Όμ„μ˜ λͺ‡ 가지 μ˜ˆμ‹œμž…λ‹ˆλ‹€:

데이터 μ„ΈνŠΈ μƒ˜ν”Œ 이미지

  • λͺ¨μžμ΄ν¬ 이미지: 이 μ΄λ―Έμ§€λŠ” λͺ¨μžμ΄ν¬λœ 데이터 μ„ΈνŠΈ μ΄λ―Έμ§€λ‘œ κ΅¬μ„±λœ ν›ˆλ ¨ 배치의 μ˜ˆμ‹œμž…λ‹ˆλ‹€. λͺ¨μžμ΄ν¬λŠ” μ—¬λŸ¬ 이미지λ₯Ό ν•˜λ‚˜μ˜ μ΄λ―Έμ§€λ‘œ κ²°ν•©ν•˜μ—¬ 각 ν›ˆλ ¨ 배치 λ‚΄μ—μ„œ λ‹€μ–‘ν•œ κ°œμ²΄μ™€ μž₯면을 늘리기 μœ„ν•΄ ν›ˆλ ¨ 쀑에 μ‚¬μš©λ˜λŠ” κΈ°μˆ μž…λ‹ˆλ‹€. 이λ₯Ό 톡해 λ‹€μ–‘ν•œ 객체 크기, μ’…νš‘λΉ„ 및 μ»¨ν…μŠ€νŠΈμ— μΌλ°˜ν™”ν•˜λŠ” λͺ¨λΈμ˜ λŠ₯λ ₯을 ν–₯μƒμ‹œν‚¬ 수 μžˆμŠ΅λ‹ˆλ‹€.

이 μ˜ˆλŠ” COCO8-Seg 데이터 μ„ΈνŠΈμ— ν¬ν•¨λœ μ΄λ―Έμ§€μ˜ λ‹€μ–‘μ„±κ³Ό λ³΅μž‘μ„±, 그리고 ν•™μŠ΅ κ³Όμ •μ—μ„œ λͺ¨μžμ΄ν¬ μ‚¬μš©μ˜ 이점을 λ³΄μ—¬μ€λ‹ˆλ‹€.

인용 및 감사

연ꡬ λ˜λŠ” 개발 μž‘μ—…μ— COCO 데이터셋을 μ‚¬μš©ν•˜λŠ” 경우 λ‹€μŒ 논문을 μΈμš©ν•΄ μ£Όμ„Έμš”:

@misc{lin2015microsoft,
      title={Microsoft COCO: Common Objects in Context},
      author={Tsung-Yi Lin and Michael Maire and Serge Belongie and Lubomir Bourdev and Ross Girshick and James Hays and Pietro Perona and Deva Ramanan and C. Lawrence Zitnick and Piotr DollΓ‘r},
      year={2015},
      eprint={1405.0312},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}

We would like to acknowledge the COCO Consortium for creating and maintaining this valuable resource for the computer vision community. For more information about the COCO dataset and its creators, visit the COCO dataset website.

자주 λ¬»λŠ” 질문

What is the COCO8-Seg dataset, and how is it used in Ultralytics YOLO11?

The COCO8-Seg dataset is a compact instance segmentation dataset by Ultralytics, consisting of the first 8 images from the COCO train 2017 setβ€”4 images for training and 4 for validation. This dataset is tailored for testing and debugging segmentation models or experimenting with new detection methods. It is particularly useful with Ultralytics YOLO11 and HUB for rapid iteration and pipeline error-checking before scaling to larger datasets. For detailed usage, refer to the model Training page.

How can I train a YOLO11n-seg model using the COCO8-Seg dataset?

To train a YOLO11n-seg model on the COCO8-Seg dataset for 100 epochs with an image size of 640, you can use Python or CLI commands. Here's a quick example:

μ—΄μ°¨ μ˜ˆμ‹œ

from ultralytics import YOLO

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

# Train the model
results = model.train(data="coco8-seg.yaml", epochs=100, imgsz=640)
# Start training from a pretrained *.pt model
yolo segment train data=coco8-seg.yaml model=yolo11n-seg.pt epochs=100 imgsz=640

μ‚¬μš© κ°€λŠ₯ν•œ 인수 및 ꡬ성 μ˜΅μ…˜μ— λŒ€ν•œ μžμ„Έν•œ μ„€λͺ…은 ꡐ윑 λ¬Έμ„œλ₯Ό μ°Έμ‘°ν•˜μ„Έμš”.

COCO8-Seg 데이터 μ„ΈνŠΈκ°€ λͺ¨λΈ 개발과 디버깅에 μ€‘μš”ν•œ μ΄μœ λŠ” λ¬΄μ—‡μΈκ°€μš”?

COCO8-Seg 데이터 μ„ΈνŠΈλŠ” μž‘μ€ ν¬κΈ°μ—μ„œ 관리성과 닀양성을 μ œκ³΅ν•˜λŠ” 데 μ΄μƒμ μž…λ‹ˆλ‹€. 단 8개의 μ΄λ―Έμ§€λ‘œ κ΅¬μ„±λ˜μ–΄ μžˆμ–΄ λŒ€κ·œλͺ¨ 데이터 μ„ΈνŠΈμ˜ μ˜€λ²„ν—€λ“œ 없이 μ„ΈλΆ„ν™” λͺ¨λΈμ΄λ‚˜ μƒˆλ‘œμš΄ 탐지 접근법을 λΉ λ₯΄κ²Œ ν…ŒμŠ€νŠΈν•˜κ³  디버깅할 수 μžˆμŠ΅λ‹ˆλ‹€. λ”°λΌμ„œ λŒ€κ·œλͺ¨ 데이터 μ„ΈνŠΈμ— λŒ€ν•œ κ΄‘λ²”μœ„ν•œ ν•™μŠ΅μ„ μ§„ν–‰ν•˜κΈ° 전에 건전성 검사 및 νŒŒμ΄ν”„λΌμΈ 였λ₯˜ 식별을 μœ„ν•œ 효율적인 λ„κ΅¬λ‘œ μ‚¬μš©ν•  수 μžˆμŠ΅λ‹ˆλ‹€. μ—¬κΈ°μ—μ„œ 데이터 μ„ΈνŠΈ ν˜•μ‹μ— λŒ€ν•΄ μžμ„Ένžˆ μ•Œμ•„λ³΄μ„Έμš”.

COCO8-Seg 데이터 μ„ΈνŠΈμ˜ YAML ꡬ성 νŒŒμΌμ€ μ–΄λ””μ—μ„œ 찾을 수 μžˆλ‚˜μš”?

COCO8-Seg 데이터 μ„ΈνŠΈμ— λŒ€ν•œ YAML ꡬ성 νŒŒμΌμ€ Ultralytics λ¦¬ν¬μ§€ν† λ¦¬μ—μ„œ 확인할 수 μžˆμŠ΅λ‹ˆλ‹€. μ—¬κΈ°μ—μ„œ 직접 νŒŒμΌμ— μ•‘μ„ΈμŠ€ν•  수 μžˆμŠ΅λ‹ˆλ‹€. YAML νŒŒμΌμ—λŠ” λͺ¨λΈ ν•™μŠ΅ 및 μœ νš¨μ„± 검사에 ν•„μš”ν•œ 데이터 μ„ΈνŠΈ 경둜, 클래슀 및 ꡬ성 섀정에 λŒ€ν•œ ν•„μˆ˜ 정보가 ν¬ν•¨λ˜μ–΄ μžˆμŠ΅λ‹ˆλ‹€.

COCO8-Seg 데이터 μ„ΈνŠΈλ‘œ ν›ˆλ ¨ν•  λ•Œ λͺ¨μžμ΄ν‚Ήμ„ μ‚¬μš©ν•˜λ©΄ μ–΄λ–€ 이점이 μžˆλ‚˜μš”?

Using mosaicing during training helps increase the diversity and variety of objects and scenes in each training batch. This technique combines multiple images into a single composite image, enhancing the model's ability to generalize to different object sizes, aspect ratios, and contexts within the scene. Mosaicing is beneficial for improving a model's robustness and accuracy, especially when working with small datasets like COCO8-Seg. For an example of mosaiced images, see the Sample Images and Annotations section.


πŸ“… Created 11 months ago ✏️ Updated 8 days ago

λŒ“κΈ€