caltech101

  • תיאור :

Caltech-101 מורכב מתמונות של אובייקטים השייכים ל-101 מחלקות, בתוספת מחלקה אחת background clutter . כל תמונה מסומנת עם אובייקט בודד. כל מחלקה מכילה בערך 40 עד 800 תמונות, בסך הכל בסביבות 9k תמונות. התמונות הן בגדלים משתנים, עם אורכי קצה טיפוסיים של 200-300 פיקסלים. גרסה זו מכילה תוויות ברמת התמונה בלבד. מערך הנתונים המקורי מכיל גם תיבות תוחמות.

לְפַצֵל דוגמאות
'test' 6,084
'train' 3,060
  • מבנה תכונה :
FeaturesDict({
    'image': Image(shape=(None, None, 3), dtype=uint8),
    'image/file_name': Text(shape=(), dtype=string),
    'label': ClassLabel(shape=(), dtype=int64, num_classes=102),
})
  • תיעוד תכונה :
תכונה מעמד צוּרָה Dtype תיאור
FeaturesDict
תמונה תמונה (אין, אין, 3) uint8
image/file_name טֶקסט חוּט
תווית ClassLabel int64

רְאִיָה

  • ציטוט :
@article{FeiFei2004LearningGV,
  title={Learning Generative Visual Models from Few Training Examples: An Incremental Bayesian Approach Tested on 101 Object Categories},
  author={Li Fei-Fei and Rob Fergus and Pietro Perona},
  journal={Computer Vision and Pattern Recognition Workshop},
  year={2004},
}