# MobileViTv3: Mobile-Friendly Vision Transformer with Simple and Effective Fusion of Local, Global and Input Features

arXiv: https://arxiv.org/abs/2209.15159 · HTML: https://mlsanity.com/paper/2209.15159/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- Builds on 45 of its 50 references that are in the map.
- First posted on arXiv 30 Sep 2022.

## Builds on (most cited first, 30 of 45)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [Squeeze-and-Excitation Networks](https://mlsanity.com/paper/1709.01507/) (2017)
- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [SSD: Single Shot MultiBox Detector](https://mlsanity.com/paper/1512.02325/) (2015)
- [Rethinking Atrous Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1706.05587/) (2017)
- [Searching for MobileNetV3](https://mlsanity.com/paper/1905.02244/) (2019)
- [Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions](https://mlsanity.com/paper/2102.12122/) (2021)
- [ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design](https://mlsanity.com/paper/1807.11164/) (2018)
- [Tokens-to-Token ViT: Training Vision Transformers from Scratch on ImageNet](https://mlsanity.com/paper/2101.11986/) (2021)
- [Semantic Understanding of Scenes through the ADE20K Dataset](https://mlsanity.com/paper/1608.05442/) (2016)
- [SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size](https://mlsanity.com/paper/1602.07360/) (2016)
- [Designing Network Design Spaces](https://mlsanity.com/paper/2003.13678/) (2020)
- [CvT: Introducing Convolutions to Vision Transformers](https://mlsanity.com/paper/2103.15808/) (2021)
- [Multiscale Vision Transformers](https://mlsanity.com/paper/2104.11227/) (2021)
- [CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classification](https://mlsanity.com/paper/2103.14899/) (2021)
- [PVT v2: Improved Baselines with Pyramid Vision Transformer](https://mlsanity.com/paper/2106.13797/) (2021)
- [MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer](https://mlsanity.com/paper/2110.02178/) (2021)
- [CoAtNet: Marrying Convolution and Attention for All Data Sizes](https://mlsanity.com/paper/2106.04803/) (2021)
- [MViTv2: Improved Multiscale Vision Transformers for Classification and Detection](https://mlsanity.com/paper/2112.01526/) (2021)
- [Conditional Positional Encodings for Vision Transformers](https://mlsanity.com/paper/2102.10882/) (2021)
- [Early Convolutions Help Transformers See Better](https://mlsanity.com/paper/2106.14881/) (2021)

## Built on by (most cited first, 30 of 44)

- [EVA-02: A Visual Representation for Neon Genesis](https://mlsanity.com/paper/2303.11331/) (2023)
- [MambaOut: Do We Really Need Mamba for Vision?](https://mlsanity.com/paper/2405.07992/) (2024)
- [Run, Don't Walk: Chasing Higher FLOPS for Faster Neural Networks](https://mlsanity.com/paper/2303.03667/) (2023)
- [RepViT: Revisiting Mobile CNN From ViT Perspective](https://mlsanity.com/paper/2307.09283/) (2023)
- [Scaling Spike-driven Transformer with Efficient Spike Firing Approximation Training](https://mlsanity.com/paper/2411.16061/) (2024)
- [Rethinking Mobile Block for Efficient Attention-based Models](https://mlsanity.com/paper/2301.01146/) (2023)
- [TinyMIM: An Empirical Study of Distilling MIM Pre-trained Models](https://mlsanity.com/paper/2301.01296/) (2023)
- [MobileMamba: Lightweight Multi-Receptive Visual Mamba Network](https://mlsanity.com/paper/2411.15941/) (2024)
- [ElasticViT: Conflict-aware Supernet Training for Deploying Fast Vision Transformer on Diverse Mobile Devices](https://mlsanity.com/paper/2303.09730/) (2023)
- [Distilled Large Language Model-Driven Dynamic Sparse Expert Activation Mechanism](https://mlsanity.com/paper/2603.26735/) (2026)
- [HSViT: Horizontally Scalable Vision Transformer](https://mlsanity.com/paper/2404.05196/) (2024)
- [Efficient Partitioning Vision Transformer on Edge Devices for Distributed Inference](https://mlsanity.com/paper/2410.11650/) (2024)
- [EMOv2: Pushing 5M Vision Model Frontier](https://mlsanity.com/paper/2412.06674/) (2024)
- [RingMo-lite: A Remote Sensing Multi-task Lightweight Network with CNN-Transformer Hybrid Framework](https://mlsanity.com/paper/2309.09003/) (2023)
- [An Experimental Study on Exploring Strong Lightweight Vision Transformers via Masked Image Modeling Pre-Training](https://mlsanity.com/paper/2404.12210/) (2024)
- [RepNeXt: A Fast Multi-Scale CNN using Structural Reparameterization](https://mlsanity.com/paper/2406.16004/) (2024)
- [FR-Net:A Light-weight FFT Residual Net For Gaze Estimation](https://mlsanity.com/paper/2305.11875/) (2023)
- [EffLoc: Lightweight Vision Transformer for Efficient 6-DOF Camera Relocalization](https://mlsanity.com/paper/2402.13537/) (2024)
- [Improving Accuracy and Generalization for Efficient Visual Tracking](https://mlsanity.com/paper/2411.18855/) (2024)
- [Light-Weight Vision Transformer with Parallel Local and Global Self-Attention](https://mlsanity.com/paper/2307.09120/) (2023)
- [MCAD: Multi-teacher Cross-modal Alignment Distillation for efficient image-text retrieval](https://mlsanity.com/paper/2310.19654/) (2023)
- [LeYOLO, New Embedded Architecture for Object Detection](https://mlsanity.com/paper/2406.14239/) (2024)
- [Beyond Isolated Heads: Multi-Overlapped-Head Self-Attention for Vision Transformers](https://mlsanity.com/paper/2410.14874/) (2024)
- [COSNet: A Novel Semantic Segmentation Network using Enhanced Boundaries in Cluttered Scenes](https://mlsanity.com/paper/2410.24139/) (2024)
- [EM-Net: Gaze Estimation with Expectation Maximization Algorithm](https://mlsanity.com/paper/2412.08074/) (2024)
- [Distilling Textual Priors from LLM to Efficient Image Fusion](https://mlsanity.com/paper/2504.07029/) (2025)
- [Geo-RepNet: Geometry-Aware Representation Learning for Surgical Phase Recognition in Endoscopic Submucosal Dissection](https://mlsanity.com/paper/2507.09294/) (2025)
- [Exploring Superposition and Interference in State-of-the-Art Low-Parameter Vision Models](https://mlsanity.com/paper/2507.15798/) (2025)
- [Accelerating Vision Foundation Models with Drop-in Depthwise Convolution](https://mlsanity.com/paper/2605.22132/) (2026)
- [A Lightweight Feature Fusion Architecture For Resource-Constrained Crowd Counting](https://mlsanity.com/paper/2401.05968/) (2024)

## Abstract (arXiv)

MobileViT (MobileViTv1) combines convolutional neural networks (CNNs) and vision transformers (ViTs) to create light-weight models for mobile vision tasks. Though the main MobileViTv1-block helps to achieve competitive state-of-the-art results, the fusion block inside MobileViTv1-block, creates scaling challenges and has a complex learning task. We propose changes to the fusion block that are simple and effective to create MobileViTv3-block, which addresses the scaling and simplifies the learning task. Our proposed MobileViTv3-block used to create MobileViTv3-XXS, XS and S models outperform MobileViTv1 on ImageNet-1k, ADE20K, COCO and PascalVOC2012 datasets. On ImageNet-1K, MobileViTv3-XXS and MobileViTv3-XS surpasses MobileViTv1-XXS and MobileViTv1-XS by 2% and 1.9% respectively. Recently published MobileViTv2 architecture removes fusion block and uses linear complexity transformers to perform better than MobileViTv1. We add our proposed fusion block to MobileViTv2 to create MobileViTv3-0.5, 0.75 and 1.0 models. These new models give better accuracy numbers on ImageNet-1k, ADE20K, COCO and PascalVOC2012 datasets as compared to MobileViTv2. MobileViTv3-0.5 and MobileViTv3-0.75 outperforms MobileViTv2-0.5 and MobileViTv2-0.75 by 2.1% and 1.0% respectively on ImageNet-1K dataset. For segmentation task, MobileViTv3-1.0 achieves 2.07% and 1.1% better mIOU compared to MobileViTv2-1.0 on ADE20K dataset and PascalVOC2012 dataset respectively. Our code and the trained models are available at: https://github.com/micronDLA/MobileViTv3