A feature embedding strategy for high-level CNN representations from multiple convnets
A feature embedding strategy for high-level CNN representations from multiple convnets
Recently, pre-trained deep convolutional neural networks (DCNNs or ConvNets) have proven that the high-level features extracted at the top fully connected (FC) layer can improve the accuracy of various image understanding, recognition, and classification tasks. However, it has not been explored if such high-level features from different DCNN architectures contain …