Resolution Switchable Networks for Runtime Efficient Image Recognition
Tsinghua University（清华大学）；Beijing National Research Center for Information Science and Technology (BNRist)（北京信息科学与技术国家研究中心（BNRist））；Intel Labs China（英特尔中国研究院）
We propose a general method to train a single convolutional neural network which is capable of switching image resolutions at inference. Thus the running speed can be selected to meet various computational resource limits. Networks trained with the proposed method are named Resolution Switchable Networks (RS-Nets). The basic training framework shares network parameters for handling images which differ in resolution, yet keeps separate batch normalization layers. Though it is parameter-efficient in design, it leads to inconsistent accuracy variations at different resolutions, for which we provide a detailed analysis from the aspect of the train-test recognition discrepancy. A multi-resolution ensemble distillation is further designed, where a teacher is learnt on the fly as a weighted ensemble over resolutions. Thanks to the ensemble and knowledge distillation, RS-Nets enjoy accuracy improvements at a wide range of resolutions compared with individually trained models. Extensive experiments on the ImageNet dataset are provided, and we additionally consider quantization problems. Code and models are available at https://github.com/yikaiw/RS-Nets.
Monash University（莫纳什大学）；The University of Adelaide（阿德莱德大学）
Instantaneous and on demand accuracy-efficiency trade-off has been recently explored in the context of neural networks slimming. In this paper, we propose a flexible quantization strategy, termed Switchable Precision neural Networks (SP-Nets), to train a shared network capable of operating at multiple quantization levels. At runtime, the network can adjust its precision on the fly according to instant memory, latency, power consumption and accuracy demands. For example, by constraining the network weights to 1-bit with switchable precision activations, our shared network spans from BinaryConne…
Visual pattern recognition over agricultural areas is an important application of aerial image processing. In this paper, we consider the multi-modality nature of agricultural aerial images and show that naively combining different modalities together without taking the feature divergence into account can lead to sub-optimal results. Thus, we apply a Switchable Normalization block to our DeepLabV3 segmentation model to alleviate the feature divergence. Using the popular symmetric Kullback Leibler divergence measure, we show that our model can greatly reduce the divergence between RGB and near-…
Recently, a massive number of deep learning based approaches have been successfully applied to various remote sensing image (RSI) recognition tasks. However, most existing advances of deep learning methods in the RSI field heavily rely on the features extracted by the manually designed backbone network, which severely hinders the potential of deep learning models due the complexity of RSI and the limitation of prior knowledge. In this paper, we research a new design paradigm for the backbone architecture in RSI recognition tasks, including scene classification, land-cover classification and ob…
Attention mechanism has gained great success in vision recognition. Many works are devoted to improving the effectiveness of attention mechanism, which finely design the structure of the attention operator. These works need lots of experiments to pick out the optimal settings when scenarios change, which consumes a lot of time and computational resources. In addition, a neural network often contains many network layers, and most studies often use the same attention module to enhance different network layers, which hinders the further improvement of the performance of the self-attention mechani…
Understanding driver activity is vital for in-vehicle systems that aim to reduce the incidence of car accidents rooted in cognitive distraction. Automating real-time behavior recognition while ensuring actions classification with high accuracy is however challenging, given the multitude of circumstances surrounding drivers, the unique traits of individuals, and the computational constraints imposed by in-vehicle embedded platforms. Prior work fails to jointly meet these runtime/accuracy requirements and mostly rely on a single sensing modality, which in turn can be a single point of failure. I…
Deep neural networks based methods have been proved to achieve outstanding performance on object detection and classification tasks. Despite significant performance improvement, due to the deep structures, they still require prohibitive runtime to process images and maintain the highest possible performance for real-time applications. Observing the phenomenon that human vision system (HVS) relies heavily on the temporal dependencies among frames from the visual input to conduct recognition efficiently, we propose a novel framework dubbed as TKD: temporal knowledge distillation. This framework …
Conventional model quantization methods use a fixed quantization scheme to different data samples, which ignores the inherent "recognition difficulty" differences between various samples. We propose to feed different data samples with varying quantization schemes to achieve a data-dependent dynamic inference, at a fine-grained layer level. However, enabling this adaptive inference with changeable layer-wise quantization schemes is challenging because the combination of bit-widths and layers is growing exponentially, making it extremely difficult to train a single model in such a vast searching…
The Hong Kong University of Science and Technology（香港科技大学）；Intel Labs China（英特尔中国研究院）
Convolutional Neural Networks (CNNs) do not have a predictable recognition behavior with respect to the input resolution change. This prevents the feasibility of deployment on different input image resolutions for a specific model. To achieve efficient and flexible image classification at runtime, we employ meta learners to generate convolutional weights of main networks for various input scales and maintain privatized Batch Normalization layers per scale. For improved training performance, we further utilize knowledge distillation on the fly over model predictions based on different input res…
Predicting human fixations from images has recently seen large improvements by leveraging deep representations which were pretrained for object recognition. However, as we show in this paper, these networks are highly overparameterized for the task of fixation prediction. We first present a simple yet principled greedy pruning method which we call Fisher pruning. Through a combination of knowledge distillation and Fisher pruning, we obtain much more runtime-efficient architectures for saliency prediction, achieving a 10x speedup for the same AUC performance as a state of the art network on the…
Carnegie Mellon University（卡内基梅隆大学）；University of California San Diego（加利福尼亚大学圣迭戈分校）；Shanghai University of Finance and Economics（上海财经大学）；University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Generic Image recognition is a fundamental and fairly important visual problem in computer vision. One of the major challenges of this task lies in the fact that single image usually has multiple objects inside while the labels are still one-hot, another one is noisy and sometimes missing labels when annotated by humans. In this paper, we focus on tackling these challenges accompanying with two different image recognition problems: multi-model ensemble and noisy data recognition with a unified framework. As is well-known, usually the best performing deep neural models are ensembles of multiple…
Convolutional Neural Network (CNN) recognition rates drop in the presence of noise. We demonstrate a novel method of counteracting this drop in recognition rate by adjusting the biases of the neurons in the convolutional layers according to the noise conditions encountered at runtime. We compare our technique to training one network for all possible noise levels, dehazing via preprocessing a signal with a denoising autoencoder, and training a network specifically for each noise level. Our system compares favorably in terms of robustness, computational complexity and recognition rate.
University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）；Snap Inc.（斯纳普公司）；ByteDance Inc.（字节跳动公司）
We present a simple and general method to train a single neural network executable at different widths (number of channels in a layer), permitting instant and adaptive accuracy-efficiency trade-offs at runtime. Instead of training individual networks with different width configurations, we train a shared network with switchable batch normalization. At runtime, the network can adjust its width on the fly according to on-device benchmarks and resource constraints, rather than downloading and offloading different models. Our trained networks, named slimmable neural networks, achieve similar (and …
Convolutional Neural Networks (CNN) are very popular in many fields including computer vision, speech recognition, natural language processing, to name a few. Though deep learning leads to groundbreaking performance in these domains, the networks used are very demanding computationally and are far from real-time even on a GPU, which is not power efficient and therefore does not suit low power systems such as mobile devices. To overcome this challenge, some solutions have been proposed for quantizing the weights and activations of these networks, which accelerate the runtime significantly. Yet,…
We present the Teacher-Student Generative Adversarial Network (TS-GAN) to generate depth images from single RGB images in order to boost the performance of face recognition systems. For our method to generalize well across unseen datasets, we design two components in the architecture, a teacher and a student. The teacher, which itself consists of a generator and a discriminator, learns a latent mapping between input RGB and paired depth images in a supervised fashion. The student, which consists of two generators (one shared with the teacher) and a discriminator, learns from new RGB data with …
If the training dataset is not very large, image recognition is usually implemented with the transfer learning methods. In these methods the features are extracted using a deep convolutional neural network, which was preliminarily trained with an external very-large dataset. In this paper we consider the nonparametric classification of extracted feature vectors with the probabilistic neural network (PNN). The number of neurons at the pattern layer of the PNN is equal to the database size, which causes the low recognition performance and high memory space complexity of this network. We propose …
University of Essex（埃塞克斯大学）；Queensland University of Technology（昆士兰科技大学）
Visual place recognition (VPR) is a robot's ability to determine whether a place was visited before using visual data. While conventional hand-crafted methods for VPR fail under extreme environmental appearance changes, those based on convolutional neural networks (CNNs) achieve state-of-the-art performance but result in heavy runtime processes and model sizes that demand a large amount of memory. Hence, CNN-based approaches are unsuitable for resource-constrained platforms, such as small robots and drones. In this paper, we take a multi-step approach of decreasing the precision of model param…
This paper presents VTN, a transformer-based framework for video recognition. Inspired by recent developments in vision transformers, we ditch the standard approach in video action recognition that relies on 3D ConvNets and introduce a method that classifies actions by attending to the entire video sequence information. Our approach is generic and builds on top of any given 2D spatial network. In terms of wall runtime, it trains $16.1\times$ faster and runs $5.1\times$ faster during inference while maintaining competitive accuracy compared to other state-of-the-art methods. It enables whole vi…
To increase the computational efficiency of interest-point based object retrieval, researchers have put remarkable research efforts into improving the efficiency of kNN-based feature matching, pursuing to match thousands of features against a database within fractions of a second. However, due to the high-dimensional nature of image features that reduces the effectivity of index structures (curse of dimensionality), due to the vast amount of features stored in image databases (images are often represented by up to several thousand features), this ultimate goal demanded to trade query runtimes …
Nanjing University of Science and Technology（南京理工大学）
Point cloud based retrieval for place recognition is still a challenging problem due to drastic appearance and illumination changes of scenes in changing environments. Existing deep learning based global descriptors for the retrieval task usually consume a large amount of computation resources (e.g., memory), which may not be suitable for the cases of limited hardware resources. In this paper, we develop an efficient point cloud learning network (EPC-Net) to form a global descriptor for visual place recognition, which can obtain good performance and reduce computation memory and inference time…
Georgia Institute of Technology（佐治亚理工学院）；Indiana University（印第安纳大学）；EgoVid Inc.（EgoVid公司）
The prevalence of Internet of things (IoT) devices and abundance of sensor data has created an increase in real-time data processing such as recognition of speech, image, and video. While currently such processes are offloaded to the computationally powerful cloud system, a localized and distributed approach is desirable because (i) it preserves the privacy of users and (ii) it omits the dependency on cloud services. However, IoT networks are usually composed of resource-constrained devices, and a single device is not powerful enough to process real-time data. To overcome this challenge, we ex…
Huazhong University of Science and Technology（华中科技大学）
Scene text detection is an important step of scene text recognition system and also a challenging problem. Different from general object detection, the main challenges of scene text detection lie on arbitrary orientations, small sizes, and significantly variant aspect ratios of text in natural images. In this paper, we present an end-to-end trainable fast scene text detector, named TextBoxes++, which detects arbitrary-oriented scene text with both high accuracy and efficiency in a single network forward pass. No post-processing other than an efficient non-maximum suppression is involved. We ha…
SenseTime Research（商汤科技研究院）；The Chinese University of Hong Kong (Shenzhen)（香港中文大学（深圳））；Shenzhen Research Institute of Big Data（深圳大数据研究院）；The University of Hong Kong（香港大学）
Normalization techniques are important in different advanced neural networks and different tasks. This work investigates a novel dynamic learning-to-normalize (L2N) problem by proposing Exemplar Normalization (EN), which is able to learn different normalization methods for different convolutional layers and image samples of a deep network. EN significantly improves flexibility of the recently proposed switchable normalization (SN), which solves a static L2N problem by linearly combining several normalizers in each normalization layer (the combination is the same for all samples). Instead of di…
University of Essex（埃塞克斯大学）；Queensland University of Technology（昆士兰科技大学）；University of Southampton（南安普顿大学）
Visual place recognition (VPR) is an essential component of robot navigation and localization systems that allows them to identify a place using only image data. VPR is challenging due to the significant changes in a place's appearance under different illumination throughout the day, with seasonal weather and when observed from different viewpoints. Currently, no single VPR technique excels in every environmental condition, each exhibiting unique benefits and shortcomings. As a result, VPR systems combining multiple techniques achieve more reliable VPR performance in changing environments, at …
Towards Seamless Adaptation of Pre-trained Models for Visual Place Recognition
Tsinghua Shenzhen International Graduate School, Tsinghua University（清华大学深圳国际研究生院）；Peng Cheng Laboratory（鹏城实验室）；University of Chinese Academy of Sciences（中国科学院大学）
Due to the limits of bandwidth and storage space, digital images are usually down-scaled and compressed when transmitted over networks, resulting in loss of details and jarring artifacts that can lower the performance of high-level visual tasks. In this paper, we aim to generate an artifact-free high-resolution image from a low-resolution one compressed with an arbitrary quality factor by exploring joint compression artifacts reduction (CAR) and super-resolution (SR) tasks. First, we propose a context-aware joint CAR and SR neural network (CAJNN) that integrates both local and non-local featur…
