คาลเทค101

  • คำอธิบาย :

Caltech-101 ประกอบด้วยรูปภาพของวัตถุที่อยู่ในคลาส 101 และคลาส background clutter หนึ่งคลาส แต่ละภาพจะมีป้ายกำกับด้วยวัตถุชิ้นเดียว แต่ละชั้นเรียนมีรูปภาพประมาณ 40 ถึง 800 ภาพ รวมประมาณ 9,000 ภาพ รูปภาพมีหลายขนาด โดยมีความยาวขอบโดยทั่วไปอยู่ที่ 200-300 พิกเซล เวอร์ชันนี้มีป้ายกำกับระดับรูปภาพเท่านั้น ชุดข้อมูลดั้งเดิมยังมีกรอบล้อมรอบด้วย

แยก ตัวอย่าง
'test' 6,084
'train' 3,060
  • โครงสร้างคุณสมบัติ :
FeaturesDict({
    'image': Image(shape=(None, None, 3), dtype=uint8),
    'image/file_name': Text(shape=(), dtype=string),
    'label': ClassLabel(shape=(), dtype=int64, num_classes=102),
})
  • เอกสารคุณสมบัติ :
คุณสมบัติ ระดับ รูปร่าง ประเภทD คำอธิบาย
คุณสมบัติDict
ภาพ ภาพ (ไม่มี ไม่มี 3) uint8
รูปภาพ/file_name ข้อความ เชือก
ฉลาก ClassLabel int64

การแสดงภาพ

  • การอ้างอิง :
@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},
}