CIFAR-10包含10個類別,50,000個訓練圖像,彩色圖像大小:32x32,10,000個測試圖像。php
(類別:airplane,automobile, bird, cat, deer, dog, frog, horse, ship, truck)html
(做者:Alex Krizhevsky, Vinod Nair, and Geoffrey Hinton)git
(數據格式:Python版本、Matlab版本、二進制版本<for C程序>)github
CIFAR-100與CIFAR-10相似,包含100個類,每類有600張圖片,其中500張用於訓練,100張用於測試;這100個類分組成20個超類。每一個圖像有一個"find" label和一個"coarse"label。測試
圖像分類結果及應的論文,包含數據集:MNIST、CIFAR-十、CIFAR-100、STL-十、SVHN、ILSVRC2012 task 1 ui
ILSVRC: ImageNet Large Scale Visual Recognition Challenge人工智能
ImageNet相關信息以下:.net
1)Total number of non-empty synsets: 21841
2)Total number of images: 14,197,122
3)Number of images with bounding box annotations: 1,034,908
4)Number of synsets with SIFT features: 1000
5)Number of images with SIFT features: 1.2 millioncode
COCO(Common Objects in Context)是一個新的圖像識別、分割、和字幕數據集,它有以下特色:orm
1)Object segmentation
2)Recognition in Context
3)Multiple objects per image
4)More than 300,000 images
5)More than 2 Million instances
6)80 object categories
7)5 captions per image
8)Keypoints on 100,000 people
COCO 2016 Detection Challenge(2016.6.1-2016.9.9)和COCO 2016 Keypoint Challenge(2016.6.1-2016.9.9)已經由Microsoft發起 由ECCV 2016(ECCV:European Conference On Computer Vision )。
3)Human3.6M (3D Human Pose Dataset)
- 《Iterated Second-Order Label Sensitive Pooling for 3D Human Pose Estimation》
1)LFW (Labeled Faces in the Wild)
3)KITTI Vision Benchmark Suite