MobileNetV2 is a classification model (distinct from MobileNetSSDv2) developed by Google. It provides real-time classification capabilities under computing constraints in devices like smartphones. This implementation leverages transfer learning from ImageNet to your dataset.
The MobileNetV2 architecture utilizes an inverted residual structure where the input and output of the residual blocks are thin bottleneck layers. MobileNetV2 also uses lightweight convolutions to filter features in the expansion layer. Finally, it removes non-linearities in the narrow layers.
Image in Courtesy of Papers With Code
MobileNet V2 outperforms MobileNet V1 with higher accuracies and lower latencies.
Image in Courtesy of Google AI
How to Train MobileNetV2 On a Custom Dataset: https://blog.roboflow.com/how-to-train-mobilenetv2-on-a-custom-dataset/
MobileNetV2 Paper: https://arxiv.org/abs/1801.04381v4