A survey on Image Data Augmentation for Deep Learning
Citations
3,940 citations
1,084 citations
884 citations
Cites background from "A survey on Image Data Augmentation..."
...In almost all areas of deep learning [40], dataset augmentation is the standard solution against overfitting....
[...]
845 citations
806 citations
References
123,388 citations
73,978 citations
55,235 citations
"A survey on Image Data Augmentation..." refers methods in this paper
...This has led to a sequence of progressively more complex architectures from AlexNet [1] to VGG-16 [2], ResNet [3], Inception-V3 [4], and DenseNet [5]....
[...]
...This differs from Transfer Learning because in Transfer Learning, the network architecture such as VGG-16 [2] or ResNet [3] must be transferred as well as the weights....
[...]
...The history of Deep Learning advancement from feature engineering such as SIFT [113] and HOG [114] to architecture design such as AlexNet [1], VGGNet [2], and Inception-V3 [4], suggest that meta-architecture design is the next paradigm shift....
[...]
49,914 citations
49,639 citations
"A survey on Image Data Augmentation..." refers methods in this paper
...[59] found a ~ 3% classification accuracy drop between grayscale and RGB images with their experiments on ImageNet [12] and the PASCAL [60] VOC dataset....
[...]
...In Pretraining, the network architecture is defined and then trained on a big dataset such as ImageNet [12]....
[...]
...Transfer Learning works by training a network on a big dataset such as ImageNet [12] and then using those weights as the initial weights in a new classification task....
[...]