smallnorb

ฐานข้อมูลนี้มีไว้สำหรับการทดลองในการจดจำวัตถุ 3 มิติจากรูปร่าง ประกอบด้วยรูปภาพของเล่น 50 ชิ้นใน 5 หมวดหมู่ทั่วไป ได้แก่ สัตว์สี่ขา หุ่นคน เครื่องบิน รถบรรทุก และรถยนต์ วัตถุถูกถ่ายภาพโดยกล้องสองตัวภายใต้สภาพแสง 6 แบบ ระดับความสูง 9 ระดับ (30 ถึง 70 องศาทุกๆ 5 องศา) และ 18 แนวราบ (0 ถึง 340 ทุกๆ 20 องศา)

ชุดฝึกประกอบด้วย 5 ตัวอย่างของแต่ละหมวด (ชุดที่ 4, 6, 7, 8 และ 9) และชุดทดสอบที่เหลืออีก 5 ชุด (ชุดที่ 0, 1, 2, 3 และ 5)

แยก ตัวอย่าง
'test' 24,300
'train' 24,300
  • โครงสร้างคุณลักษณะ :
FeaturesDict({
    'image': Image(shape=(96, 96, 1), dtype=uint8),
    'image2': Image(shape=(96, 96, 1), dtype=uint8),
    'instance': ClassLabel(shape=(), dtype=int64, num_classes=10),
    'label_azimuth': ClassLabel(shape=(), dtype=int64, num_classes=18),
    'label_category': ClassLabel(shape=(), dtype=int64, num_classes=5),
    'label_elevation': ClassLabel(shape=(), dtype=int64, num_classes=9),
    'label_lighting': ClassLabel(shape=(), dtype=int64, num_classes=6),
})
  • เอกสารคุณสมบัติ :
คุณสมบัติ ระดับ รูปร่าง Dประเภท คำอธิบาย
คุณสมบัติDict
ภาพ ภาพ (96, 96, 1) uint8
ภาพที่ 2 ภาพ (96, 96, 1) uint8
ตัวอย่าง ป้ายกำกับคลาส int64
label_azimuth ป้ายกำกับคลาส int64
label_category ป้ายกำกับคลาส int64
label_elevation ป้ายกำกับคลาส int64
label_lighting ป้ายกำกับคลาส int64
  • การอ้างอิง :
@article{LeCun2004LearningMF,
  title={Learning methods for generic object recognition with invariance to pose and lighting},
  author={Yann LeCun and Fu Jie Huang and L{\'e}on Bottou},
  journal={Proceedings of the 2004 IEEE Computer Society Conference on Computer Vision and Pattern Recognition},
  year={2004},
  volume={2},
  pages={II-104 Vol.2}
}