PaperCVPR 2018 (arXiv 2018) · 2018
MobileNetV2: Inverted Residuals and Linear Bottlenecks
1page built on it
2more pages cite it
1narrative
1registry paper cites it
In the Atlas
ModelPrimary source of
MobileNetV2
Efficient mobile CNN backbone built from inverted-residual blocks with a linear bottleneck — depthwise-separable convolution expanded to a wide interior and projected back to a thin, non-linearity-free bottleneck that the residual connects — for on-device classification, detection (SSDLite), and segmentation (Mobile DeepLabv3).
Cited as a reference by 2 pages, grouped by domain
Features 2
Appears in 1 narrative
Lineage
Papers in the Atlas registry only · one dot per paperBuilt upon by · 1
Builds on · 4
2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksRen, He et al.2015Going deeper with convolutionsSzegedy, Liu et al.2016Deep Residual Learning for Image RecognitionHe, Zhang et al.2018DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFsChen, Papandreou et al.