mlsanityData to 7 Oct 2026

MobileNetV2: Inverted Residuals and Linear Bottlenecks

arXiv 1801.04381 ↗First posted 13 Jan 2018cs.CVcited by 4,773

Not on the map. Why?

Cited by 50 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

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

Key facts. Cited by 4,773 papers in the map (data through 7 Oct 2026). First posted on arXiv 13 Jan 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2018: 1; 2019: 0; 2020: 1; 2021: 1036; 2022: 929; 2023: 893; 2024: 765; 2025: 733; 2026: 415.

4,773 papers.

Sort:

By area:

  1. 2022A ConvNet for the 2020scited by 3,545
  2. 2021EfficientNetV2: Smaller Models and Faster Trainingcited by 1,127
  3. 2021Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNetcited by 1,096
  4. 2022YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectorscited by 978
  5. 2021Uformer: A General U-Shaped Transformer for Image Restorationcited by 864
  6. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 798
  7. 2021MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformercited by 636
  8. 2021ImageNet-21K Pretraining for the Massescited by 607
  9. 2021CoAtNet: Marrying Convolution and Attention for All Data Sizescited by 566
  10. 2021Twins: Revisiting the Design of Spatial Attention in Vision Transformerscited by 565
  11. 2021CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windowscited by 526
  12. 2021RepVGG: Making VGG-style ConvNets Great Againcited by 520
38 more
  1. 2024YOLOv10: Real-Time End-to-End Object Detectioncited by 508
  2. 2021ReAct: Out-of-distribution Detection With Rectified Activationscited by 486
  3. 2022MaxViT: Multi-Axis Vision Transformercited by 476
  4. 2021BRECQ: Pushing the Limit of Post-Training Quantization by Block Reconstructioncited by 464
  5. 2021A White Paper on Neural Network Quantizationcited by 453
  6. 2022PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategiescited by 432
  7. 2022Decoupled Knowledge Distillationcited by 380
  8. 2021Pay Attention to MLPscited by 367
  9. 2021Rethinking Spatial Dimensions of Vision Transformerscited by 361
  10. 2021High-Performance Large-Scale Image Recognition Without Normalizationcited by 359
  11. 2023Faster Segment Anything: Towards Lightweight SAM for Mobile Applicationscited by 345
  12. 2021A Field Guide to Federated Optimizationcited by 312
  13. 2021MERLOT: Multimodal Neural Script Knowledge Modelscited by 311
  14. 2021CMT: Convolutional Neural Networks Meet Vision Transformerscited by 310
  15. 2021Distilling Knowledge via Knowledge Reviewcited by 285
  16. 2021Escaping the Big Data Paradigm with Compact Transformerscited by 261
  17. 2021I-BERT: Integer-only BERT Quantizationcited by 250
  18. 2021Accuracy on the Line: On the Strong Correlation Between Out-of-Distribution and In-Distribution Generalizationcited by 248
  19. 2022Transformers in Medical Imaging: A Surveycited by 245
  20. 2021No Fear of Heterogeneity: Classifier Calibration for Federated Learning with Non-IID Datacited by 241
  21. 2022EfficientFormer: Vision Transformers at MobileNet Speedcited by 240
  22. 2021Coordinate Attention for Efficient Mobile Network Designcited by 239
  23. 2021Adaptive Fourier Neural Operators: Efficient Token Mixers for Transformerscited by 237
  24. 2024LocalMamba: Visual State Space Model with Windowed Selective Scancited by 235
  25. 2021Mobile-Former: Bridging MobileNet and Transformercited by 234
  26. 2021LocalViT: Analyzing Locality in Vision Transformerscited by 228
  27. 2021AdaViT: Adaptive Vision Transformers for Efficient Image Recognitioncited by 228
  28. 2021Associating Objects with Transformers for Video Object Segmentationcited by 225
  29. 2022CDDFuse: Correlation-Driven Dual-Branch Feature Decomposition for Multi-Modality Image Fusioncited by 222
  30. 2021ViTAE: Vision Transformer Advanced by Exploring Intrinsic Inductive Biascited by 216
  31. 2021RobustNet: Improving Domain Generalization in Urban-Scene Segmentation via Instance Selective Whiteningcited by 213
  32. 2021Accelerating Sparse Deep Neural Networkscited by 213
  33. 2022Extremely Simple Activation Shaping for Out-of-Distribution Detectioncited by 212
  34. 2024Diffusion Model-Based Image Editing: A Surveycited by 212
  35. 2022Visual Attention Networkcited by 210
  36. 2021Fast Yet Effective Machine Unlearningcited by 209
  37. 2021WildDeepfake: A Challenging Real-World Dataset for Deepfake Detectioncited by 207
  38. 2022UniFormer: Unifying Convolution and Self-attention for Visual Recognitioncited by 206

50 most cited of 4,773 All 4,773 (JSON)