Cluster Pruning: An Efficient Filter Pruning Method for Edge AI Vision Applications

Abstract

Even though the Convolutional Neural Networks (CNN) has shown superior results in the field of computer vision, it is still a challenging task to implement computer vision algorithms in real-time at the edge, especially using a low-cost IoT device due to high memory consumption and computation complexities in a CNN. Network compression methodologies such as weight pruning, filter pruning, and quantization are used to overcome the above mentioned problem. Even though filter pruning methodology has shown better performances compared to other techniques, irregularity of the number of filters pruned across different layers of a CNN might not comply with majority of the neural computing hardware architectures. In this paper, a novel greedy approach called cluster pruning has been proposed, which provides a structured way of removing filters in a CNN by considering the importance of filters and the underlying hardware architecture. The proposed methodology is compared with the conventional filter pruning algorithm on Pascal-VOC open dataset, and Head-Counting dataset, which is our own dataset developed to detect and count people entering a room. We benchmark our proposed method on three hardware architectures, namely CPU, GPU, and Intel Movidius Neural Computer Stick (NCS) using the popular SSD-MobileNet and SSD-SqueezeNet neural network architectures used for edge-AI vision applications. Results demonstrate that our method outperforms the conventional filter pruning methodology, using both datasets on above mentioned hardware architectures. Furthermore, a low cost IoT hardware setup consisting of an Intel Movidius-NCS is proposed to deploy an edge-AI application using our proposed pruning methodology.

References in corpus (18)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• SSD: Single Shot MultiBox Detector
• MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications
• Deep Compression: Compressing Deep Neural Networks with Pruning, Trained Quantization and Huffman Coding
• Pruning Convolutional Neural Networks for Resource Efficient Inference
• AMC: AutoML for Model Compression and Acceleration on Mobile Devices
• Rethinking the Value of Network Pruning
• Network Trimming: A Data-Driven Neuron Pruning Approach towards Efficient Deep Architectures
• A Survey of Data Fusion in Smart City Applications
• BigDL: A Distributed Deep Learning Framework for Big Data
• EIE: Efficient Inference Engine on Compressed Deep Neural Network
• SCNN: An Accelerator for Compressed-sparse Convolutional Neural Networks
• N2N Learning: Network to Network Compression via Policy Gradient Reinforcement Learning
• Tiny-DSOD: Lightweight Object Detection for Resource-Restricted Usages
• Network Decoupling: From Regular to Depthwise Separable Convolutions
• Deep Neural Network Compression with Single and Multiple Level Quantization
• Compact Deep Convolutional Neural Networks With Coarse Pruning
• Caffe con Troll: Shallow Ideas to Speed Up Deep Learning

Cited by in corpus (5)

• WiFi Fingerprint Clustering for Urban Mobility Analysis
• The Study of Urban Residential's Public Space Activeness using Space-centric Approach
• SmartSplit: Latency-Energy-Memory Optimisation for CNN Splitting on Smartphone Environment
• A Real-time Low-cost Artificial Intelligence System for Autonomous Spraying in Palm Plantations
• Latency-Memory Optimized Splitting of Convolution Neural Networks for Resource Constrained Edge Devices