Mobile CNN

Mobile CNNs are convolutional neural networks optimized for deployment on resource-constrained mobile devices, aiming to balance high accuracy with low computational cost and latency. Current research focuses on developing novel architectures like EfficientNet, MobileNetV2, and hybrid CNN-ViT models (e.g., RepViT, MoCoViT) that achieve state-of-the-art performance while minimizing computational demands. These advancements are significant for expanding the accessibility of AI-powered applications in mobile devices, impacting diverse fields such as smart agriculture, mobile photography, and real-time object detection. Further improvements are being explored through techniques like sparse attention mechanisms and hardware-software co-design, including the integration of in-memory computing architectures.

Papers

August 3, 2024

Advancing Green AI: Efficient and Accurate Lightweight CNNs for Rice Leaf Disease Identification
Khairun Saddami, Yudha Nurdin, Mutia Zahramita, Muhammad Shahreeza Safiruz
Smart Agriculture Green AI Lightweight CNN Mobile CNN Rice Leaf Disease Rice Disease

November 30, 2023

Corrupting Convolution-based Unlearnable Datasets with Pixel-based Image Transformations
Xianlong Wang, Shengshan Hu, Minghui Li, Zhifei Yu, Ziqi Zhou, Leo Yu Zhang
Adversarial Training Additive Noise Corruption Level Unlearnable Datasets Mobile CNN Convolution Based Unlearnable

July 18, 2023

RepViT: Revisiting Mobile CNN From ViT Perspective
Ao Wang, Hui Chen, Zijia Lin, Jungong Han, Guiguang Ding
Convolutional Neural Network Lightweight Convolutional Neural Network Lightweight CNN Mobile CNN Lightweight Vision Transformer

July 1, 2023

MobileViG: Graph-Based Sparse Attention for Mobile Vision Applications
Mustafa Munir, William Avery, Radu Marculescu
Mobile CNN Mobile Vision Vision Graph

April 18, 2023

Heterogeneous Integration of In-Memory Analog Computing Architectures with Tensor Processing Units
Mohammed E. Elbtity, Brendan Reidy, Md Hasibul Amin, Ramtin Zand
Analog in Memory Computing Tensor Processing Unit Heterogeneous System Edge TPU Mobile CNN

November 8, 2022

PyNet-V2 Mobile: Efficient On-Device Photo Processing With Neural Networks
Andrey Ignatov, Grigory Malivenko, Radu Timofte, Yu Tseng, Yu-Syuan Xu, Po-Hsiang Yu, Cheng-Ming Chiang, Hsien-Kai Kuo, Min-Hung Chen, Chia-Ming Cheng, Luc Van Gool
Neural Network Image Sensor Mobile CNN Mobile Photography Device Vision Task

July 15, 2022

Lightweight Vision Transformer with Cross Feature Attention
Youpeng Zhao, Huadong Tang, Yingying Jiang, Yong A, Qiang Wu
Convolutional Neural Network Vision Transformer Mobile CNN Cross Image Attention

May 25, 2022

MoCoViT: Mobile Convolutional Vision Transformer
Hailong Ma, Xin Xia, Xing Wang, Xuefeng Xiao, Jiashi Li, Min Zheng
Vision Transformer Mobile CNN

December 5, 2021

Boosting Mobile CNN Inference through Semantic Memory
Yun Li, Chen Zhang, Shihao Han, Li Lyna Zhang, Baoqun Yin, Yunxin Liu, Mengwei Xu
Device Deep Semantic Memory Hierarchical Memory Memory Capacity Mobile CNN