Convolution operator is the core of convolutional neural networks (CNNs) and occupies the most computation cost. To make CNNs more efficient, many methods have been proposed to either design lightweight networks or compress models. Although some efficient network structures have been proposed, such as MobileNet or ShuffleNet, we find that there still exists redundant information between convolution kernels. To address this issue, we propose a novel dynamic convolution method to adaptively generate convolution kernels based on image contents. To demonstrate the effectiveness, we apply dynamic convolution on multiple state-of-the-art CNNs. On one hand, we can reduce the computation cost remarkably while maintaining the performance. For ShuffleNetV2/MobileNetV2/ResNet18/ResNet50, DyNet can reduce 37.0/54.7/67.2/71.3% FLOPs without loss of accuracy. On the other hand, the performance can be largely boosted if the computation cost is maintained. Based on the architecture MobileNetV3-Small/Large, DyNet achieves 70.3/77.1% Top-1 accuracy on ImageNet with an improvement of 2.9/1.9%. To verify the scalability, we also apply DyNet on segmentation task, the results show that DyNet can reduce 69.3% FLOPs while maintaining Mean IoU on segmentation task.
Huazhong University of Science and Technology（华中科技大学）；City University of Hong Kong（香港城市大学）
Convolution operations dominate the overall execution time of Convolutional Neural Networks (CNNs). This paper proposes an easy yet efficient technique for both Convolutional Neural Network training and testing. The conventional convolution and pooling operations are replaced by Easy Convolution and Random Pooling (ECP). In ECP, we randomly select one pixel out of four and only conduct convolution operations of the selected pixel. As a result, only a quarter of the conventional convolution computations are needed. Experiments demonstrate that the proposed EasyConvPooling can achieve 1.45x spee…
Expanding the receptive field to capture large-scale context is key to obtaining good performance in dense prediction tasks, such as human pose estimation. While many state-of-the-art fully-convolutional architectures enlarge the receptive field by reducing resolution using strided convolution or pooling layers, the most straightforward strategy is adopting large filters. This, however, is costly because of the quadratic increase in the number of parameters and multiply-add operations. In this work, we explore using learnable box filters to allow for convolution with arbitrarily large kernel s…
University of Oulu（奥卢大学）；South China University of Technology（华南理工大学）；National University of Defense Technology（国防科技大学）
Replacing normal convolutions with group convolutions can significantly increase the computational efficiency of modern deep convolutional networks, which has been widely adopted in compact network architecture designs. However, existing group convolutions undermine the original network structures by cutting off some connections permanently resulting in significant accuracy degradation. In this paper, we propose dynamic group convolution (DGC) that adaptively selects which part of input channels to be connected within each group for individual samples on the fly. Specifically, we equip each gr…
Google Brain（谷歌大脑）；New York University（纽约大学）；Google Deepmind（谷歌DeepMind）
Efficient simulation of the Navier-Stokes equations for fluid flow is a long standing problem in applied mathematics, for which state-of-the-art methods require large compute resources. In this work, we propose a data-driven approach that leverages the approximation power of deep-learning with the precision of standard solvers to obtain fast and highly realistic simulations. Our method solves the incompressible Euler equations using the standard operator splitting method, in which a large sparse linear system with many free parameters must be solved. We use a Convolutional Network with a highl…
To address the limitations of existing magnitude-based pruning algorithms in cases where model weights or activations are of large and similar magnitude, we propose a novel perspective to discover parameter redundancy among channels and accelerate deep CNNs via channel pruning. Precisely, we argue that channels revealing similar feature information have functional overlap and that most channels within each such similarity group can be removed without compromising model's representational power. After deriving an effective metric for evaluating channel similarity through probabilistic modeling,…
The deep learning revolution brought us an extensive array of neural network architectures that achieve state-of-the-art performance in a wide variety of Computer Vision tasks including among others, classification, detection and segmentation. In parallel, we have also been observing an unprecedented demand in computational and memory requirements, rendering the efficient use of neural networks in low-powered devices virtually unattainable. Towards this end, we propose a three-stage compression and acceleration pipeline that sparsifies, quantizes and entropy encodes activation maps of Convolut…
Chinese Academy of Sciences（中国科学院）；The Johns Hopkins University（约翰斯·霍普金斯大学）；Microsoft Research（微软研究院）
This paper presents an efficient module named spatial bottleneck for accelerating the convolutional layers in deep neural networks. The core idea is to decompose convolution into two stages, which first reduce the spatial resolution of the feature map, and then restore it to the desired size. This operation decreases the sampling density in the spatial domain, which is independent yet complementary to network acceleration approaches in the channel domain. Using different sampling rates, we can tradeoff between recognition accuracy and model complexity. As a basic building block, spatial bottle…
As a successful deep model applied in image super-resolution (SR), the Super-Resolution Convolutional Neural Network (SRCNN) has demonstrated superior performance to the previous hand-crafted models either in speed and restoration quality. However, the high computational cost still hinders it from practical usage that demands real-time performance (24 fps). In this paper, we aim at accelerating the current SRCNN, and propose a compact hourglass-shape CNN structure for faster and better SR. We re-design the SRCNN structure mainly in three aspects. First, we introduce a deconvolution layer at th…
Accelerating Vision Foundation Models with Drop-in Depthwise Convolution
University of Modena and Reggio Emilia（摩德纳和雷吉奥艾米利亚大学）；INSAIT, Sofia University ``St. Kliment Ohridski'', Sofia, Bulgaria（INSAIT，索菲亚大学『圣克莱门特·欧赫里迪斯』，索菲亚，保加利亚）
Filter pruning is one of the most effective ways to accelerate and compress convolutional neural networks (CNNs). In this work, we propose a global filter pruning algorithm called Gate Decorator, which transforms a vanilla CNN module by multiplying its output by the channel-wise scaling factors, i.e. gate. When the scaling factor is set to zero, it is equivalent to removing the corresponding filter. We use Taylor expansion to estimate the change in the loss function caused by setting the scaling factor to zero and use the estimation for the global filter importance ranking. Then we prune the n…
Southern University of Science and Technology（南方科技大学）；University of Technology Sydney（悉尼科技大学）；Fudan University（复旦大学）
This paper proposed a Soft Filter Pruning (SFP) method to accelerate the inference procedure of deep Convolutional Neural Networks (CNNs). Specifically, the proposed SFP enables the pruned filters to be updated when training the model after pruning. SFP has two advantages over previous works: (1) Larger model capacity. Updating previously pruned filters provides our approach with larger optimization space than fixing the filters to zero. Therefore, the network trained by our method has a larger model capacity to learn from the training data. (2) Less dependence on the pre-trained model. Large …
Point source detection techniques are used to identify and localise point sources in radio astronomical surveys. With the development of the Square Kilometre Array (SKA) telescope, survey images will see a massive increase in size from Gigapixels to Terapixels. Point source detection has already proven to be a challenge in recent surveys performed by SKA pathfinder telescopes. This paper proposes the Point Proposal Network (PPN): a point source detector that utilises deep convolutional neural networks for fast source detection. Results measured on simulated MeerKAT images show that, although l…
University of Chinese Academy of Sciences（中国科学院大学）；Columbia University（哥伦比亚大学）
Deep unsupervised domain adaptation (UDA) has recently received increasing attention from researchers. However, existing methods are computationally intensive due to the computation cost of Convolutional Neural Networks (CNN) adopted by most work. To date, there is no effective network compression method for accelerating these models. In this paper, we propose a unified Transfer Channel Pruning (TCP) approach for accelerating UDA models. TCP is capable of compressing the deep UDA model by pruning less important channels while simultaneously learning transferable features by reducing the cross-…
Memorial Sloan Kettering Cancer Center（纪念斯隆凯特琳癌症中心）；Graz University of Technology（格拉茨工业大学）；GE Healthcare（GE医疗）
Purpose: To investigate feasibility of accelerating prostate diffusion-weighted imaging (DWI) by reducing the number of acquired averages and denoising the resulting image using a proposed guided denoising convolutional neural network (DnCNN). Materials and Methods: Raw data from the prostate DWI scans were retrospectively gathered (between July 2018 and July 2019) from six single-vendor MRI scanners. 118 data sets were used for training and validation (age: 64.3 +- 8 years) and 37 - for testing (age: 65.1 +- 7.3 years). High b-value diffusion-weighted (hb-DW) data were reconstructed into nois…
Pacific Northwest National Laboratory（太平洋西北国家实验室）；University of Texas at Arlington（德克萨斯大学阿灵顿分校）
Deep Learning is a very powerful machine learning model. Deep Learning trains a large number of parameters for multiple layers and is very slow when data is in large scale and the architecture size is large. Inspired from the shrinking technique used in accelerating computation of Support Vector Machines (SVM) algorithm and screening technique used in LASSO, we propose a shrinking Deep Learning with recall (sDLr) approach to speed up deep learning computation. We experiment shrinking Deep Learning with recall (sDLr) using Deep Neural Network (DNN), Deep Belief Network (DBN) and Convolution Neu…
Huawei Noah’s Ark Lab（华为诺亚方舟实验室）；University of Sydney（悉尼大学）
Exploring deep convolutional neural networks of high efficiency and low memory usage is very essential for a wide variety of machine learning tasks. Most of existing approaches used to accelerate deep models by manipulating parameters or filters without data, e.g., pruning and decomposition. In contrast, we study this problem from a different perspective by respecting the difference between data. An instance-wise feature pruning is developed by identifying informative features for different instances. Specifically, by investigating a feature decay regularization, we expect intermediate feature…
This paper presents cltorch, a hardware-agnostic backend for the Torch neural network framework. cltorch enables training of deep neural networks on GPUs from diverse hardware vendors, including AMD, NVIDIA, and Intel. cltorch contains sufficient implementation to run models such as AlexNet, VGG, Overfeat, and GoogleNet. It is written using the OpenCL language, a portable compute language, governed by the Khronos Group. cltorch is the top-ranked hardware-agnostic machine learning framework on Chintala's convnet-benchmarks page. This paper presents the technical challenges encountered whilst cr…
Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；University of Chinese Academy of Sciences（中国科学院大学）；Jilin University（吉林大学）
Accelerating deep convolutional neural networks has become an active topic and sparked an interest in academia and industry. In this paper, we propose an efficient low-precision quantized Winograd convolution algorithm, called LANCE, which combines the advantages of fast convolution and quantization techniques. By embedding linear quantization operations into the Winograd-domain, the fast convolution can be performed efficiently under low-precision computation on graphics processing units. We test neural network models with LANCE on representative image classification datasets, including SVHN,…
A novel neural network architecture, known as DL-ESPIRiT, is proposed to reconstruct rapidly acquired cardiac MRI data without field-of-view limitations which are present in previously proposed deep learning-based reconstruction frameworks. Additionally, a novel convolutional neural network based on separable 3D convolutions is integrated into DL-ESPIRiT to more efficiently learn spatiotemporal priors for dynamic image reconstruction. The network is trained on fully-sampled 2D cardiac cine datasets collected from eleven healthy volunteers with IRB approval. DL-ESPIRiT is compared against a sta…
Hefei University of Technology（合肥工业大学）；Institute of Computing Technology, Chinese Academy of Sciences（中国科学院计算技术研究所）；National University of Singapore（新加坡国立大学）
Generative neural network is a new category of neural networks and it has been widely utilized in applications such as content generation, unsupervised learning, segmentation and pose estimation. It typically involves massive computing-intensive deconvolution operations that cannot be fitted to conventional neural network processors directly. However, prior works mainly investigated specialized hardware architectures through intensive hardware modifications to the existing deep learning processors to accelerate deconvolution together with the convolution. In contrast, this work proposes a nove…
In this paper, we introduce a new channel pruning method to accelerate very deep convolutional neural networks.Given a trained CNN model, we propose an iterative two-step algorithm to effectively prune each layer, by a LASSO regression based channel selection and least square reconstruction. We further generalize this algorithm to multi-layer and multi-branch cases. Our method reduces the accumulated error and enhance the compatibility with various architectures. Our pruned VGG-16 achieves the state-of-the-art results by 5x speed-up along with only 0.3% increase of error. More importantly, our…
Convolutional Neural Networks (CNNs) are currently adopted to solve an ever greater number of problems, ranging from speech recognition to image classification and segmentation. The large amount of processing required by CNNs calls for dedicated and tailored hardware support methods. Moreover, CNN workloads have a streaming nature, well suited to reconfigurable hardware architectures such as FPGAs. The amount and diversity of research on the subject of CNN FPGA acceleration within the last 3 years demonstrates the tremendous industrial and academic interest. This paper presents a state-of-the-…
Convolutional neural networks (CNN) have achieved impressive performance on the wide variety of tasks (classification, detection, etc.) across multiple domains at the cost of high computational and memory requirements. Thus, leveraging CNNs for real-time applications necessitates model compression approaches that not only reduce the total number of parameters but reduce the overall computation as well. In this work, we present a stability-based approach for filter-level pruning of CNNs. We evaluate our proposed approach on different architectures (LeNet, VGG-16, ResNet, and Faster RCNN) and da…
Many real-world signal sources are complex-valued, having real and imaginary components. However, the vast majority of existing deep learning platforms and network architectures do not support the use of complex-valued data. MRI data is inherently complex-valued, so existing approaches discard the richer algebraic structure of the complex data. In this work, we investigate end-to-end complex-valued convolutional neural networks - specifically, for image reconstruction in lieu of two-channel real-valued networks. We apply this to magnetic resonance imaging reconstruction for the purpose of acce…
We introduce Spatial Group Convolution (SGC) for accelerating the computation of 3D dense prediction tasks. SGC is orthogonal to group convolution, which works on spatial dimensions rather than feature channel dimension. It divides input voxels into different groups, then conducts 3D sparse convolution on these separated groups. As only valid voxels are considered when performing convolution, computation can be significantly reduced with a slight loss of accuracy. The proposed operations are validated on semantic scene completion task, which aims to predict a complete 3D volume with semantic l…
FastOcc: Accelerating 3D Occupancy Prediction by Fusing the 2D Bird's-Eye View and Perspective View
School of Computer Science, Fudan University（复旦大学计算机科学技术学院）；Mogo Auto Intelligence and Telematics Information Technology Co., Ltd（摩国汽车智能与远程信息处理信息技术有限公司）；Faculty of Information Technology, Beijing University of Technology（北京工业大学信息学部）；Institute of Science and Technology for Brain-Inspired Intelligence, Fudan University（复旦大学脑智能科学与技术研究院）
High bandwidth requirements are an obstacle for accelerating the training and inference of deep neural networks. Most previous research focuses on reducing the size of kernel maps for inference. We analyze parameter sparsity of six popular convolutional neural networks - AlexNet, MobileNet, ResNet-50, SqueezeNet, TinyNet, and VGG16. Of the networks considered, those using ReLU (AlexNet, SqueezeNet, VGG16) contain a high percentage of 0-valued parameters and can be statically pruned. Networks with Non-ReLU activation functions in some cases may not contain any 0-valued parameters (ResNet-50, Ti…
Deep image prior (DIP) is a recently proposed technique for solving imaging inverse problems by fitting the reconstructed images to the output of an untrained convolutional neural network. Unlike pretrained feedforward neural networks, the same DIP can generalize to arbitrary inverse problems, from denoising to phase retrieval, while offering competitive performance at each task. The central disadvantage of DIP is that, while feedforward neural networks can reconstruct an image in a single pass, DIP must gradually update its weights over hundreds to thousands of iterations, at a significant co…
High-Frequency Prior-Driven Adaptive Masking for Accelerating Image Super-Resolution
School of Computer Science and Technology, Harbin Institute of Technology（计算机科学与技术学院，哈尔滨工业大学）；City University of Hong Kong（香港城市大学）；Tianjin University（天津大学）
Southern University of Science and Technology（南方科技大学）；Shanghai Aircraft Design and Research Institute（上海飞机设计研究院）
For the goal of automated design of high-performance deep convolutional neural networks (CNNs), Neural Architecture Search (NAS) methodology is becoming increasingly important for both academia and industries.Due to the costly stochastic gradient descent (SGD) training of CNNs for performance evaluation, most existing NAS methods are computationally expensive for real-world deployments. To address this issue, we first introduce a new performance estimation metric, named Random-Weight Evaluation (RWE) to quantify the quality of CNNs in a cost-efficient manner. Instead of fully training the enti…
University of Southern California（南加州大学）；Samsung Research America（三星美国研究院）
Face detection is a widely studied problem over the past few decades. Recently, significant improvements have been achieved via the deep neural network, however, it is still challenging to directly apply these techniques to mobile devices for its limited computational power and memory. In this work, we present a proposal generation acceleration framework for real-time face detection. More specifically, we adopt a popular cascaded convolutional neural network (CNN) as the basis, then apply our acceleration approach on the basic framework to speed up the model inference time. We are motivated by…
A large number of retinal vessel analysis methods based on image segmentation have emerged in recent years. However, existing methods depend on cumbersome backbones, such as VGG16 and ResNet-50, benefiting from their powerful feature extraction capabilities but suffering from high computational costs. In this paper, we propose a novel neural network (HybridNetSeg) dedicated to solving this drawback while further improving overall performance. Considering deformable convolution can extract complex and variable structural information, and larger kernel in mixed depthwise convolution makes contri…
Korea Advanced Institute of Science and Technology（韩国高等科学技术院）
Low-rank decomposition plays a central role in accelerating convolutional neural network (CNN), and the rank of decomposed kernel-tensor is a key parameter that determines the complexity and accuracy of a neural network. In this paper, we define rank selection as a combinatorial optimization problem and propose a methodology to minimize network complexity while maintaining the desired accuracy. Combinatorial optimization is not feasible due to search space limitations. To restrict the search space and obtain the optimal rank, we define the space constraint parameters with a boundary condition.…
Light binary convolutional neural networks (LB-CNN) are particularly useful when implemented in low-energy computing platforms as required in many industrial applications. Herein, a framework for optimizing compact LB-CNN is introduced and its effectiveness is evaluated. The framework is freely available and may run on free-access cloud platforms, thus requiring no major investments. The optimized model is saved in the standardized .h5 format and can be used as input to specialized tools for further deployment into specific technologies, thus enabling the rapid development of various intellige…
Chubu University（中部大学）；Denso IT Laboratory（电装IT实验室）
Recent trends show recognition accuracy increasing even more profoundly. Inference process of Deep Convolutional Neural Networks (DCNN) has a large number of parameters, requires a large amount of computation, and can be very slow. The large number of parameters also require large amounts of memory. This is resulting in increasingly long computation times and large model sizes. To implement mobile and other low performance devices incorporating DCNN, model sizes must be compressed and computation must be accelerated. To that end, this paper proposes Binary-decomposed DCNN, which resolves these…
Instituto de Ciência e Tecnologia, Universidade Federal de São Paulo – UNIFESP（圣保罗联邦大学科学技术研究所（UNIFESP））
Most image data available are often stored in a compressed format, from which JPEG is the most widespread. To feed this data on a convolutional neural network (CNN), a preliminary decoding process is required to obtain RGB pixels, demanding a high computational load and memory usage. For this reason, the design of CNNs for processing JPEG compressed data has gained attention in recent years. In most existing works, typical CNN architectures are adapted to facilitate the learning with the DCT coefficients rather than RGB pixels. Although they are effective, their architectural changes either ra…
AI Institute（360人工智能研究院）；CAI, University of Technology Sydney（悉尼科技大学CAI）；National University of Singapore（新加坡国立大学）
In this paper, we present a novel and general network structure towards accelerating the inference process of convolutional neural networks, which is more complicated in network structure yet with less inference complexity. The core idea is to equip each original convolutional layer with another low-cost collaborative layer (LCCL), and the element-wise multiplication of the ReLU outputs of these two parallel layers produces the layer-wise output. The combined layer is potentially more discriminative than the original convolutional layer, and its inference is faster for two reasons: 1) the zero…
Very large overhead imagery associated with ground truth maps has the potential to generate billions of training image patches for machine learning algorithms. However, random sampling selection criteria often leads to redundant and noisy-image patches for model training. With minimal research efforts behind this challenge, the current status spells missed opportunities to develop supervised learning algorithms that generalize over wide geographical scenes. In addition, much of the computational cycles for large scale machine learning are poorly spent crunching through noisy and redundant imag…
We present an approach to accelerating a wide variety of image processing operators. Our approach uses a fully-convolutional network that is trained on input-output pairs that demonstrate the operator's action. After training, the original operator need not be run at all. The trained network operates at full resolution and runs in constant time. We investigate the effect of network architecture on approximation accuracy, runtime, and memory footprint, and identify a specific architecture that balances these considerations. We evaluate the presented approach on ten advanced image processing ope…
The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））
Filter pruning has attracted increasing attention in recent years for its capacity in compressing and accelerating convolutional neural networks. Various data-independent criteria, including norm-based and relationship-based ones, were proposed to prune the most unimportant filters. However, these state-of-the-art criteria fail to fully consider the dissimilarity of filters, and thus might lead to performance degradation. In this paper, we first analyze the limitation of relationship-based criteria with examples, and then introduce a new data-independent criterion, Weighted Hybrid Criterion (W…
LGViT: Dynamic Early Exiting for Accelerating Vision Transformer
Beijing Institute of Technology（北京理工大学）；JD Explore Academy（京东探索研究院）；Wuhan University（武汉大学）；China Academic of Electronics and Information Technology（中国电子信息技术研究院）；City, University of London（伦敦大学城市学院）
University of Tennessee（田纳西大学）；Sun Yat-Sen University（中山大学）；Oak Ridge National Laboratory（橡树岭国家实验室）
Channel-based pruning has achieved significant successes in accelerating deep convolutional neural network, whose pipeline is an iterative three-step procedure: ranking, pruning and fine-tuning. However, this iterative procedure is computationally expensive. In this study, we present a novel computationally efficient channel pruning approach based on the coarse ranking that utilizes the intermediate results during fine-tuning to rank the importance of filters, built upon state-of-the-art works with data-driven ranking criteria. The goal of this work is not to propose a single improved approach…
Beijing Laboratory of Intelligent Information Technology, School of Computer Science, Beijing Institute of Technology（北京理工大学计算机学院智能信息技术北京市重点实验室）
Recent studies have shown that aggregating convolutional features of a pre-trained Convolutional Neural Network (CNN) can obtain impressive performance for a variety of visual tasks. The symmetric Positive Definite (SPD) matrix becomes a powerful tool due to its remarkable ability to learn an appropriate statistic representation to characterize the underlying structure of visual features. In this paper, we propose to aggregate deep convolutional features into an SPD matrix representation through the SPD generation and the SPD transformation under an end-to-end deep network. To this end, severa…
Benefiting from the great success of deep learning in computer vision, CNN-based object detection methods have drawn significant attentions. Various frameworks have been proposed which show awesome and robust performance for a large range of datasets. However, for building detection in remote sensing images, buildings always pose a diversity of orientations which makes it a challenge for the application of off-the-shelf methods to building detection. In this work, we aim to integrate orientation regression into the popular axis-aligned bounding-box detection method to tackle this problem. To a…
Imperial College London（帝国理工学院）；King’s College London（伦敦国王学院）
Accelerating the data acquisition of dynamic magnetic resonance imaging (MRI) leads to a challenging ill-posed inverse problem, which has received great interest from both the signal processing and machine learning community over the last decades. The key ingredient to the problem is how to exploit the temporal correlation of the MR sequence to resolve the aliasing artefact. Traditionally, such observation led to a formulation of a non-convex optimisation problem, which were solved using iterative algorithms. Recently, however, deep learning based-approaches have gained significant popularity …
