From the Publisher: 
The accessible presentation of this book gives both a general view of the entire computer vision enterprise and also offers sufficient detail to be able to build useful applications. Users learn techniques that have proven to be useful by first-hand experience and a wide range of mathematical methods. A CD-ROM with every copy of the text contains source code for programming practice, color images, and illustrative movies. Comprehensive and up-to-date, this book includes essential topics that either reflect practical significance or are of theoretical importance. Topics are discussed in substantial and increasing depth. Application surveys describe numerous important application areas such as image based rendering and digital libraries. Many important algorithms broken down and illustrated in pseudo code. Appropriate for use by engineers as a comprehensive reference to the computer vision enterprise.

Computer vision : a modern approach = 计算机视觉 : 一种现代的方法

The current lightweight face recognition models need improvement in terms of floating point operations (FLOPs), parameters, and model size. Motivated by ConvNeXt and MobileFaceNet, a family of lightweight face recognition models known as ConvFaceNeXt is introduced to overcome the shortcomings listed above. ConvFaceNeXt has three main parts, which are the stem, bottleneck, and embedding partitions. Unlike ConvNeXt, which applies the revamped inverted bottleneck dubbed the ConvNeXt block in a large ResNet-50 model, the ConvFaceNeXt family is designed as lightweight models. The enhanced ConvNeXt (ECN) block is proposed as the main building block for ConvFaceNeXt. The ECN block contributes significantly to lowering the FLOP count. In addition to the typical downsampling approach using convolution with a kernel size of three, a patchify strategy utilizing a kernel size of two is also implemented as an alternative for the ConvFaceNeXt family. The purpose of adopting the patchify strategy is to reduce the computational complexity further. Moreover, blocks with the same output dimension in the bottleneck partition are added together for better feature correlation. Based on the experimental results, the proposed ConvFaceNeXt model achieves competitive or even better results when compared with previous lightweight face recognition models, on top of a significantly lower FLOP count, parameters, and model size.

/pdf/convfacenext-lightweight-networks-for-face-recognition-1pe9233c.pdf

ConvFaceNeXt: Lightweight Networks for Face Recognition

The superior performance of the recent deep learning models comes at the cost of a significant increase in computational complexity, memory use, and power consumption. Filter pruning is one of the effective neural network compression techniques suitable for model deployment in modern low-power edge devices. In this paper, we propose a loss-aware filter Magnitude and Similarity based Variable rate Filter Pruning (MSVFP) technique. We studied several filter selection criteria based on filter magnitude and similarity among filters within a convolution layer, and based on the assumption that the sensitivity of each layer throughout the network is different, unlike conventional fixed rate pruning methods, our algorithm using loss-aware filter selection criteria automatically finds the suitable pruning rate for each layer throughout the network. In addition, the proposed algorithm adapts two different filter selection criteria to remove weak filters as well as redundant filters based on filter magnitude and filter similarity score respectively. Finally, the iterative filter pruning and retraining approach are used to maintain the accuracy of the network during pruning to its target float point operations (FLOPs) reduction rate. In the proposed algorithm, a small number of retraining steps are sufficient during iterative pruning to prevent an abrupt drop in the accuracy of the network. Experiments with commonly used VGGNet and ResNet models on CIFAR-10 and ImageNet benchmark show the superiority of the proposed method over the existing methods in the literature. Notably, VGG-16, ResNet-56, and ResNet-110 models on the CIFAR-10 dataset even improved the original accuracy with more than 50% reduction in network FLOPs. Additionally, the ResNet-50 model on the ImageNet dataset reduces model FLOPs by more than 42% with a negligible drop in the original accuracy.

/pdf/magnitude-and-similarity-based-variable-rate-filter-pruning-2f6n8fze.pdf

Magnitude and Similarity Based Variable Rate Filter Pruning for Efficient Convolution Neural Networks

Convolutional neural network pruning based on misclassification cost

Automatic classification of aerial images has become one of the topics studied in recent years. Especially for the use of drones in different fields such as agricultural applications, smart city applications, surveillance and security applications, it is necessary to automatically classify the images obtained with the camera during autonomous mission execution. For this purpose, researchers have created new data sets and some computer vision methods have been developed to achieve high accuracy. However, in addition to increasing the accuracy of the developed methods, the computational complexity should also be reduced. Because the methods to be used in devices such as drones where energy consumption is important should have low computational complexity. In this study, firstly, five different state-of-art deep learning models were used to obtain high accuracy values in the classification of aerial images. Among these models, the VGG19 model achieved the highest accuracy with 94.21%. In the second part of the study, the parameters of this model were analyzed and the model was reconstructed. The number of 143.6 million parameters of the VGG19 model was reduced to 34 million. The accuracy of the model obtained by reducing the number of parameters is 93.56% on the same test data. Thus, despite the 66.5% decrease in the parameter ratio, there was only a 0.7% decrease in the accuracy value. When compared to previous studies, the results show improved performance.

/pdf/classification-of-scenes-in-aerial-images-with-deep-learning-3mrd47nr.pdf

Classification of Scenes in Aerial Images with Deep Learning Models

Adaptive CNN filter pruning using global importance metric

In this paper, the problem of estimating the depth of an object from its monocular image is addressed. Here, basically, an algorithm is developed, which performs shape matching, and as a result, achieve accurate depth maps of objects. In the algorithm, first, an optimal camera position is determined. Then, the 3D model is projected onto the image plane of the camera, yielding a projected 2D image of the 3D model. An objective function determines a score based on graph-based feature matching, and the depth map is extracted from the geometrical information of the 3D model. Finally, the depth map and the original image are combined to create a 3D point cloud simulation of the object. Experimental analysis shows the efficacy of the proposed method.

Bishshoy Das

Papers

Adaptive CNN filter pruning using global importance metric

Graph-Based Depth Estimation in a Monocular Image Using Constrained 3D Wireframe Models.

Knowledge Diversification in Ensembles of Identical Neural Networks

Edge-Aware Image Super-Resolution Using a Generative Adversarial Network