Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

VeloxNet: Efficient Spatial Gating for Lightweight Embedded Image Classification

Md Meftahul Ferdaus, Elias Ioup, Mahdi Abdelguerfi, Anton Netchaev, Steven Sloan, Ken Pathak, Kendall N. Niles

Abstract

Deploying deep learning models on embedded devices for tasks such as aerial disaster monitoring and infrastructure inspection requires architectures that balance accuracy with strict constraints on model size, memory, and latency. This paper introduces VeloxNet, a lightweight CNN architecture that replaces SqueezeNet's fire modules with gated multi-layer perceptron (gMLP) blocks for embedded image classification. Each gMLP block uses a spatial gating unit (SGU) that applies learned spatial projections and multiplicative gating, enabling the network to capture spatial dependencies across the full feature map in a single layer. Unlike fire modules, which are limited to local receptive fields defined by small convolutional kernels, the SGU provides global spatial modeling at each layer with fewer parameters. We evaluate VeloxNet on three aerial image datasets: the Aerial Image Database for Emergency Response (AIDER), the Comprehensive Disaster Dataset (CDD), and the Levee Defect Dataset (LDD), comparing against eleven baselines including MobileNet variants, ShuffleNet, EfficientNet, and recent vision transformers. VeloxNet reduces the parameter count by 46.1% relative to SqueezeNet (from 740,970 to 399,366) while improving weighted F1 scores by 6.32% on AIDER, 30.83% on CDD, and 2.51% on LDD. These results demonstrate that substituting local convolutional modules with spatial gating blocks can improve both classification accuracy and parameter efficiency for resource-constrained deployment. The source code will be made publicly available upon acceptance of the paper.

VeloxNet: Efficient Spatial Gating for Lightweight Embedded Image Classification

Abstract

Deploying deep learning models on embedded devices for tasks such as aerial disaster monitoring and infrastructure inspection requires architectures that balance accuracy with strict constraints on model size, memory, and latency. This paper introduces VeloxNet, a lightweight CNN architecture that replaces SqueezeNet's fire modules with gated multi-layer perceptron (gMLP) blocks for embedded image classification. Each gMLP block uses a spatial gating unit (SGU) that applies learned spatial projections and multiplicative gating, enabling the network to capture spatial dependencies across the full feature map in a single layer. Unlike fire modules, which are limited to local receptive fields defined by small convolutional kernels, the SGU provides global spatial modeling at each layer with fewer parameters. We evaluate VeloxNet on three aerial image datasets: the Aerial Image Database for Emergency Response (AIDER), the Comprehensive Disaster Dataset (CDD), and the Levee Defect Dataset (LDD), comparing against eleven baselines including MobileNet variants, ShuffleNet, EfficientNet, and recent vision transformers. VeloxNet reduces the parameter count by 46.1% relative to SqueezeNet (from 740,970 to 399,366) while improving weighted F1 scores by 6.32% on AIDER, 30.83% on CDD, and 2.51% on LDD. These results demonstrate that substituting local convolutional modules with spatial gating blocks can improve both classification accuracy and parameter efficiency for resource-constrained deployment. The source code will be made publicly available upon acceptance of the paper.
Paper Structure (19 sections, 9 equations, 2 figures, 6 tables)

This paper contains 19 sections, 9 equations, 2 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Problem Formulation
  3. Notation and Preliminaries
  4. Resource-Constrained Image Classification
  5. The Efficiency-Accuracy Trade-off
  6. Spatial Gating as an Alternative Inductive Bias
  7. Design Objective of VeloxNet
  8. VeloxNet: Bridging gMLP Theory and Lightweight CNN Practice
  9. Motivation and Novel Architectural Contributions
  10. VeloxNet Architecture Design
  11. Efficient Spatial Gating Unit (SGU)
  12. Architectural Details and Fire Module Replacement Strategy
  13. Comparative Architectural Analysis: SqueezeNet vs. VeloxNet
  14. Results and Discussion
  15. Experimental Setup
  16. ...and 4 more sections

Figures (2)

  • Figure 1: Macro-architectural view of the original SqueezeNet architecture along with micro-architectural view showing organization of convolution filters in the fire module
  • Figure 2: Macro-architectural view of the proposed VeloxNet architecture along with micro-architectural view of the gMLP architecture featuring our efficient SGU implementation