MobileNetV2: Inverted Residuals and Linear Bottlenecks
read the original abstract
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. Additionally, we demonstrate how to build mobile semantic segmentation models through a reduced form of DeepLabv3 which we call Mobile DeepLabv3. The MobileNetV2 architecture is based on an inverted residual structure where the input and output of the residual block are thin bottleneck layers opposite to traditional residual models which use expanded representations in the input an MobileNetV2 uses lightweight depthwise convolutions to filter features in the intermediate expansion layer. Additionally, we find that it is important to remove non-linearities in the narrow layers in order to maintain representational power. We demonstrate that this improves performance and provide an intuition that led to this design. Finally, our approach allows decoupling of the input/output domains from the expressiveness of the transformation, which provides a convenient framework for further analysis. We measure our performance on Imagenet classification, COCO object detection, VOC image segmentation. We evaluate the trade-offs between accuracy, and number of operations measured by multiply-adds (MAdd), as well as the number of parameters
This paper has not been read by Pith yet.
Forward citations
Cited by 11 Pith papers
-
Towards Real-Time ECG and EMG Modeling on $\mu$NPUs
PhysioLite delivers Transformer-comparable ECG/EMG performance using learnable wavelet filters and hardware-aware design at ~370KB quantized size on μNPUs.
-
Scaling Robot Learning with Semantically Imagined Experience
Augmenting robot datasets via diffusion-based semantic inpainting enables manipulation policies to solve unseen tasks with new objects and improves robustness to novel distractors.
-
Memory- and Communication-Aware Model Compression for Distributed Deep Learning Inference on IoT
NoNN partitions a teacher model into disjoint compressed students via network science for distributed IoT inference, matching teacher accuracy with far lower per-device memory and communication.
-
Open DNN Box by Power Side-Channel Attack
Power side-channel analysis recovers DNN architecture and parameters at 96.5% average accuracy on real embedded devices.
-
EPNAS: Efficient Progressive Neural Architecture Search
EPNAS uses a progressive search policy with REINFORCE performance prediction to search neural architectures in parallel, supporting multiple resource constraints and outperforming ENAS and PNAS on CIFAR-10 and ImageNe...
-
New pointwise convolution in Deep Neural Networks through Extremely Fast and Non Parametric Transforms
Replacing pointwise convolutions with DWHT yields a model with 79.1% fewer parameters, 48.4% fewer FLOPs, and 1.49% higher accuracy than MobileNet-V1 on CIFAR-100.
-
Layer-wise Derivative Controlled Networks
ChainzRule with DREG regularization claims 15.5x fewer parameters than standard models, 23.1% lower peak gradient volatility on MNIST, and 70.17% accuracy on Yelp Full ordinal regression.
-
TeaLeafVision: An Explainable and Robust Deep Learning Framework for Tea Leaf Disease Classification
DenseNet201 CNN reaches 99% accuracy on teaLeafBD dataset for classifying tea leaf diseases, with added Grad-CAM explainability and adversarial robustness.
-
A Quantitative Evaluation of Approximate Softmax Functions for Deep Neural Networks
First- and second-order Taylor approximations to softmax deliver up to 14% FPGA resource reduction with at most 0.2% accuracy loss on LeNet-5 and MobileNet v2.
-
Genetic Network Architecture Search
Genetic algorithm searches convolution cell architectures with weight sharing via SGD, reporting 96% accuracy on CIFAR10 and 80.1% on CIFAR100.
-
Evaluation of Winning Solutions of 2025 Low Power Computer Vision Challenge
The 2025 LPCVC winners demonstrate practical techniques for low-power image classification under varied conditions, open-vocabulary segmentation from text prompts, and monocular depth estimation.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.