Deep Neural Network Compression for Image Classification and Object Detection

TL;DR

This paper introduces a novel, network-agnostic compression technique using dynamical clustering to significantly reduce the size and computational requirements of deep neural networks for image classification and object detection, enabling deployment on resource-constrained devices.

Contribution

The authors propose a new compression method that is model-agnostic and employs dynamical clustering, achieving high parameter reduction without substantial accuracy loss.

Findings

Pruned about 95% of parameters in classification networks.

Reduced model size by up to 59.70% in object detection networks.

Achieved 110X less memory usage with minimal accuracy impact.

Abstract

Neural networks have been notorious for being computationally expensive. This is mainly because neural networks are often over-parametrized and most likely have redundant nodes or layers as they are getting deeper and wider. Their demand for hardware resources prohibits their extensive use in embedded devices and puts restrictions on tasks like real-time image classification or object detection. In this work, we propose a network-agnostic model compression method infused with a novel dynamical clustering approach to reduce the computational cost and memory footprint of deep neural networks. We evaluated our new compression method on five different state-of-the-art image classification and object detection networks. In classification networks, we pruned about 95% of network parameters. In advanced detection networks such as YOLOv3, our proposed compression method managed to reduce the model parameters up to 59.70% which yielded 110X less memory without sacrificing much in accuracy.