Multi-Layer Pruning Framework for Compressing Single Shot MultiBox Detector

Abstract

We propose a framework for compressing state-of-the-art Single Shot MultiBox Detector (SSD). The framework addresses compression in the following stages: Sparsity Induction, Filter Selection, and Filter Pruning. In the Sparsity Induction stage, the object detector model is sparsified via an improved global threshold. In Filter Selection & Pruning stage, we select and remove filters using sparsity statistics of filter weights in two consecutive convolutional layers. This results in the model with the size smaller than most existing compact architectures. We evaluate the performance of our framework with multiple datasets and compare over multiple methods. Experimental results show that our method achieves state-of-the-art compression of 6.7X and 4.9X on PASCAL VOC dataset on models SSD300 and SSD512 respectively. We further show that the method produces maximum compression of 26X with SSD512 on German Traffic Sign Detection Benchmark (GTSDB). Additionally, we also empirically show our method's adaptability for classification based architecture VGG16 on datasets CIFAR and German Traffic Sign Recognition Benchmark (GTSRB) achieving a compression rate of 125X and 200X with the reduction in flops by 90.50% and 96.6% respectively with no loss of accuracy. In addition to this, our method does not require any special libraries or hardware support for the resulting compressed models.

Real-time human detection for electricity conservation using pruned-SSD and arduino

“…The results of various pruning techniques in terms of accuracy, compression rate and floating-point operations (FLOPs) are represented in Table 4. Our model out-performs the training speed of the model created by multi-layer pruning framework [41], as the latter implements three stages of pruning, only after completely training the network from the scratch. Hence the training time increases many folds in this method, whereas the former (our model) trains the network initially based on OATM method which reduces the training time drastically.…”

LdsConv: Learned Depthwise Separable Convolutions by Group Pruning

“…Refs. [30][31][32][33][34][35][36] prune filters that have a minimal contribution in the model. After removing these filters, the model is usually fine-tuned to maintain its performance.…”

“…Model compression is considered as another reliable and economic method to improve the efficiency of the convolutional neural network, which can be roughly divided into three categories: (a) Connection pruning [28,29]; (b) Filter pruning [30][31][32][33][34][35][36]; and (c) Quantization [28,[37][38][39]. These methods can effectively reduce the computation of the convolutional neural network, but this is always achieved at the price of sacrificing the accuracy.…”

An Accident Analysis and Prevention Model based on Heterogeneous Convolution Kernel

“…Many common models also use these convolutions to explore new architectures that can reduce FLOPS. Another common method to improve model efficiency is compression model, which can be roughly divided into three categories: connection pruning, filter pruning [1][2][3] and quantization [4]. There are two different goals for using efficient convolution filters.…”