MobileNetV2: Inverted Residuals and Linear Bottlenecks

CVPR 2018 Mark SandlerAndrew HowardMenglong ZhuAndrey ZhmoginovLiang-Chieh Chen

In this paper we describe a new mobile architecture, MobileNetV2, that improves the state of the art performance of mobile models on multiple tasks and benchmarks as well as across a spectrum of different model sizes. We also describe efficient ways of applying these mobile models to object detection in a novel framework we call SSDLite... (read more)

PDF Abstract

Code


↳ Quickstart in
64,120
See all 62 implementations

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT LEADERBOARD
Panoptic Segmentation COCO panoptic MobileNetV2 PQ 35.2 # 5
Image Classification ImageNet MobileNetV2 (1.4) Top 1 Accuracy 74.7% # 127