Search

Vehicle type recognition (VTR) is a quite common requirement and one of the key challenges in real surveillance scenarios, such as intelligent traffic and unmanned driving. Usually coarse-grained and fine-grained VTRs are applied in different applications, and the challenge from multiple viewpoints is critical for both cases. In this paper, we propose a feedback-enhancement multi-branch CNN (FM-CNN) to solve the challenge in these two cases. The proposed FM-CNN takes three derivatives of an image as input and leverages the advantages of hierarchical details, feedback enhancement, model average, and stronger robustness to translation and mirroring. A single global cross-entropy loss is insufficient to train such a complex CNN and so we add extra branch losses to enhance feedbacks to each branch. Though reusing pre-trained parameters, we propose a novel parameter update method to adapt FM-CNN to task-specific local visual patterns and global information in new datasets. To test the effectiveness of FM-CNN, we create our own multi-view VTR (MVVTR) data set since there are no such data sets available. And, for fine-grained VTR, we use the CompCars data set. Compared with state-of-the-art classification solutions without special preprocessing, the proposed FM-CNN demonstrates better performance in both coarse-grained and fine-grained scenarios. For coarse-grained VTR, it achieves 94.9% Top-1 accuracy on the MVVTR data set. For fine-grained VTR, it achieves 91.0% Top-1 and 97.8% Top-5 accuracies on the CompCars data set. [ABSTRACT FROM AUTHOR]

This paper presents a novel approach for video-based person re-identification using multiple convolutional neural networks (CNNs). Unlike the previous work, we intend to extract a compact yet discriminative appearance representation from several frames rather than the whole sequence. Specifically, given a video, the representative frames are selected based on the walking profile of consecutive frames. A multiple CNN architecture incorporated with feature pooling is proposed to learn and compile the features of the selected representative frames into a compact description about the pedestrian for identification. Experiments are conducted on benchmark data sets to demonstrate the superiority of the proposed method over existing person re-identification approaches. [ABSTRACT FROM AUTHOR]

Low level image restoration is an integral component of modern artificial intelligence (AI) driven camera pipelines. Most of these frameworks are based on deep neural networks which present a massive computational overhead on resource constrained platform like a mobile phone. In this paper, we propose several lightweight low-level modules which can be used to create a computationally low cost variant of a given baseline model. Recent works for efficient neural networks design have mainly focused on classification. However, low-level image processing falls under the ‘image-to-image’ translation genre which requires some additional computational modules not present in classification. This paper seeks to bridge this gap by designing generic efficient modules which can replace essential components used in contemporary deep learning based image restoration networks. We also present and analyse our results highlighting the drawbacks of applying depthwise separable convolutional kernel (a popular method for efficient classification network) for sub-pixel convolution based upsampling (a popular upsampling strategy for low-level vision applications). This shows that concepts from domain of classification cannot always be seamlessly integrated into ‘image-to-image’ translation tasks. We extensively validate our findings on three popular tasks of image inpainting, denoising and super-resolution. Our results show that proposed networks consistently output visually similar reconstructions compared to full capacity baselines with significant reduction of parameters, memory footprint and execution speeds on contemporary mobile devices. [ABSTRACT FROM AUTHOR]

Recognition and localization models can be generally decomposed into three components: encoder, decoder, and task head. In this paper, we rethink the necessity of decoder, as we observe that it brings additional computational and parametric burden. We thus propose to remove the decoder and present a bidirectional cross-scale architecture that is able to obtain rich semantic information and precise localization in a unified backbone. Extensive experiments demonstrate that, different from common encoder-decoder models and other down-sampling and up-sampling backbones, the proposed BiCSNet achieves improved performances compared to existing architectures for pixel-level tasks. In object detection, our BiCSNet brings significant performance improvement by ~ 3% AP at various scales with 13% – 23% fewer FLOPS, compared with ResNet-FPN models on COCO dataset. In Instance segmentation, the AP can be improved by 1% over SpineNet. BiCSNet is also promising for semantic segmentation tasks, as the proposed BiCSNet pre-trained on ImageNet alone significantly outperforms DeepLabv3 pre-trained on both ImageNet and COCO dataset by 1.3% in mIOU with 89% fewer FLOPs on PASCAL VOC 2012. [ABSTRACT FROM AUTHOR]

In both H.264/AVC and HEVC, the angular prediction is adopted for intra coding, which only exploits the spatial correlation between the current block and its neighboring single line reference. This angular prediction can handle the main directional patterns well, however, lacks the ability to deal with other directions. In this paper, a multi-scale convolutional neural network based intra prediction is proposed to address this problem. Specifically, a predicted block is first generated by the angular prediction, then fed into the proposed network with its neighboring reconstructed $L$ -shape to generate a more accurate predicted block. On one hand, the $L$ -shape of multiple lines provides more reliable reconstructed pixels and more contextual information to get better prediction; on the other hand, the multi-scale feature extraction takes the advantage of the feature maps in different scales to further enhance the prediction. With this multi-scale structure, the $L$ -shape can be used to refine both left-above and right-bottom pixels in the predicted block during the convolution operation. Experimental results demonstrate that compared with HEVC reference software HM 16.9, the proposed intra prediction can achieve an average of 3.4% (up to 5.6%) bitrate saving with all intra configuration. [ABSTRACT FROM AUTHOR]

In recent years, the image and video coding technologies have advanced by leaps and bounds. However, due to the popularization of image and video acquisition devices, the growth rate of image and video data is far beyond the improvement of the compression ratio. In particular, it has been widely recognized that there are increasing challenges of pursuing further coding performance improvement within the traditional hybrid coding framework. Deep convolution neural network which makes the neural network resurge in recent years and has achieved great success in both artificial intelligent and signal processing fields, also provides a novel and promising solution for image and video compression. In this paper, we provide a systematic, comprehensive and up-to-date review of neural network-based image and video compression techniques. The evolution and development of neural network-based compression methodologies are introduced for images and video respectively. More specifically, the cutting-edge video coding techniques by leveraging deep learning and HEVC framework are presented and discussed, which promote the state-of-the-art video coding performance substantially. Moreover, the end-to-end image and video coding frameworks based on neural networks are also reviewed, revealing interesting explorations on next generation image and video coding frameworks/standards. The most significant research works on the image and video coding related topics using neural networks are highlighted, and future trends are also envisioned. In particular, the joint compression on semantic and visual information is tentatively explored to formulate high efficiency signal representation structure for both human vision and machine vision, which are the two dominant signal receptors in the age of artificial intelligence. [ABSTRACT FROM AUTHOR]

Real-time detection of inshore ships plays an essential role in the efficient monitoring and management of maritime traffic and transportation for port management. Current ship detection methods which are mainly based on remote sensing images or radar images hardly meet real-time requirement due to the timeliness of image acquisition. In this paper, we propose to use visual images captured by an on-land surveillance camera network to achieve real-time detection. However, due to the complex background of visual images and the diversity of ship categories, the existing convolution neural network (CNN) based methods are either inaccurate or slow. To achieve high detection accuracy and real-time performance simultaneously, we propose a saliency-aware CNN framework for ship detection, comprising comprehensive ship discriminative features, such as deep feature, saliency map, and coastline prior. This model uses CNN to predict the category and the position of ships and uses the global contrast based salient region detection to correct the location. We also extract coastline information and respectively incorporate it into CNN and saliency detection to obtain more accurate ship locations. We implement our model on Darknet under CUDA 8.0 and CUDNN V5 and use a real-world visual image dataset for training and evaluation. The experimental results show that our model outperforms representative counterparts (Faster R-CNN, SSD, and YOLOv2) in terms of accuracy and speed. [ABSTRACT FROM AUTHOR]

Siamese networks are prevalent in person re-identification (re-id) tasks to address the similarity and dissimilarity among video frames. It mainly focuses on the inter-video variation between spatio-temporal features extracted from different videos, while the variation between features of the same video has been rarely discussed. In this paper, we introduce the concept of “mean-body” and define an intra-video loss to address the variation between spatio-temporal features of the same video. A novel loss is presented to boost the training of the re-id networks by combining the proposed intra-video loss and the Siamese loss. Specifically, the intra-video loss uses the unique mean-body of each camera viewpoint to make the video sequence more clustered, while the Siamese loss is to make the wrong matching videos more separated. To train the whole network, we update the network and the mean-body in an iterative manner. As a result, the proposed loss is expected to improve the generalization capability of the re-id networks on the testing set. Extensive results demonstrate that the presented approach outperforms the state-of-the-art algorithms on the publicly available data sets, such as PRID2011, iLIDS-VID, and MARS, in terms of re-id accuracy. [ABSTRACT FROM AUTHOR]

In this paper, we present novel sharp attention networks by adaptively sampling feature maps from convolutional neural networks for person re-identification (re-ID) problems. Due to the introduction of sampling-based attention models, the proposed approach can adaptively generate sharper attention-aware feature masks. This greatly differs from the gating-based attention mechanism that relies on soft gating functions to select the relevant features for person re-ID. In contrast, the proposed sampling-based attention mechanism allows us to effectively trim irrelevant features by enforcing the resultant feature masks to focus on the most discriminative features. It can produce sharper attentions that is more assertive in localizing subtle features relevant to re-identifying people across cameras. For this purpose, a differentiable Gumbel-Softmax sampler is employed to approximate the Bernoulli sampling to train the sharp attention networks. Extensive experimental evaluations demonstrate the superiority of this new sharp attention model for person re-ID over other related existing, published state-of-the-art works on three challenging benchmarks, including CUHK03, Market-1501, and DukeMTMC-reID. [ABSTRACT FROM AUTHOR]

In this paper, we present a novel hardware-friendly super-resolution (SR) method based on a convolutional neural network (CNN) and its dedicated hardware (HW) on field programmable gate array (FPGA). Although CNN-based SR methods have shown very promising results for SR, their computational complexities are prohibitive for hardware implementation. To the best of our knowledge, we are the first to implement a real-time CNN-based SR HW that upscales 2K full high-definition video to 4K ultra high-definition (UHD) video at 60 frames per second (fps). In our dedicated CNN-based SR HW, low-resolution input frames are processed line-by-line, and the number of convolutional filter parameters is reduced significantly by incorporating depth-wise separable convolutions with a residual connection. Our CNN-based SR HW incorporates a cascade of 1D convolutions having large receptive fields along horizontal lines while keeping vertical receptive fields minimal, which allows us to save required line memory space in achieving comparable SR performance against full 2D convolution operations. For efficient HW implementation, we use a simple and effective quantization method with little peak signal-to-noise ratio (PSNR) degradation. Also, we propose a compression method to efficiently store intermediate feature map data to reduce the number of line memories used in HW. Our HW implementation on the FPGA generates 4K UHD frames of higher PSNR values at 60 fps and shows better visual quality, compared with conventional CNN-based SR methods that are trained and tested in software. [ABSTRACT FROM AUTHOR]

In this paper, we tackle the problem of single image dehazing with a convolutional neural network. Within this network, we develop a multi-level fusion module to utilize both low-level and high-level features. The low-level features help to recover finer details, and the high-level features discover abstract semantics. They are complementary in the restoring of clear images. Moreover, a Residual Mixed-convolution Attention Module (RMAM) with an attention block is proposed to guide the network to focus on important features in the learning process. In this RMAM, group convolution, depth-wise convolution, and point-wise convolution are mixed, and thus it is much faster than its counterparts. With these two modules, we thus have an end-to-end network without explicitly estimating the atmospheric light intensity and the transmission map in the classical atmosphere scattering model. Both qualitative and quantitative experimental studies are carried out on public datasets including RESIDE, DCPDN-TestA, and the real-world dataset. The extensive results demonstrate both the effectiveness and efficiency of the proposed solution to single image dehazing. [ABSTRACT FROM AUTHOR]

In this paper, a novel perceptual image hashing scheme based on convolutional neural network (CNN) with multiple constraints is proposed, in which our deep hashing network learns the process of features extraction automatically according to the training target and then generates the final hash sequence. The combination of convolutional and pooling layers is to reduce the size of input image while deepening the channels. Then, we construct two pairs of constraints and integrate them into an overall constraint function through a strategy of weight allocation. In order to guarantee the robustness and discrimination of deep hashing network simultaneously, a new training method is developed to adjust the training set structure dynamically according to the changes of constraint values. Experimental results show that the proposed deep hashing network can achieve a satisfactory balance between perceptual robustnzess and discrimination while maintaining security. Based on the large-scale test set, receiver operating characteristic (ROC) curves, ${F}_{1}$ scores and equal error rate (EER) demonstrate the superiority of our scheme in terms of content authentication compared with some state-of-the-art schemes. [ABSTRACT FROM AUTHOR]

This paper presents an end-to-end learning architecture for video-based person re-identification by integrating convolutional neural networks (CNNs) and bidirectional recurrent neural networks (BRNNs). Given a video with consecutive frames, features of each frame are extracted with CNN and then are fed into the BRNN to get a final spatio-temporal representation about the video. Specifically, CNN acts as a Spatial Feature Extractor, while BRNN is expected to capture the temporal cues of sequential frames in both forward and backward directions, simultaneously. The whole network is trained end-to-end with a joint identification and verification manner. Experimental results on benchmark data sets show that the proposed model can effectively learn spatio-temporal features relevant for re-identification and outperforms existing video-based person re-identification methods. [ABSTRACT FROM AUTHOR]

Visual object tracking is a fundamental and time-critical vision task. However, most trackers such as SiamFC and CFNet missed the object movement and simply defined the searching region centered at the location of the target in the previous frame. So they tend to fail in the cases with severe occlusion or a large displacement of the target. In this paper, we consider the object tracking as a dual-task problem of guessing and matching. A guess module is to estimate the motion trend of the target by reinforcement learning based on the observations on appearance changes and motion history. Rather than using the previous location of the target, we may have a more accurate center to locate the searching region. Benefited from such improved searching region, the match module becomes less prone to the object drift problem, and can easily identify the target from the potential distractors in the background. Extensive experimental results on benchmark datasets such as RGBT, OTB-2013, OTB-50 and OTB-100, show that the proposed method achieves leading performance compared to state-of-the-art trackers. Moreover, the proposed tracker could maintain real-time speed, giving itself the potential in practical applications. [ABSTRACT FROM AUTHOR]

Deep learning provides a great potential for in-loop filtering to improve both coding efficiency and subjective quality in video coding. State-of-the-art work focuses on network structure design and employs a single powerful network to solve all problems. In contrast, this paper proposes a deep learning based systematic approach that includes an effective Convolutional Neural Network (CNN) structure, a hierarchical training strategy, and a video codec oriented switchable mechanism. First, we propose a novel CNN structure, i.e., Squeeze-and-Excitation Filtering CNN (SEFCNN), as an optional in-loop filter. To capture the non-linear interaction between channels, the SEFCNN is comprised of two subnets, i.e., Feature EXtracting (FEX) subnet and Feature ENhancing (FEN) subnet. Then, we develop a hierarchical model training strategy to adapt the two subnets to different coding scenarios. For high-rate videos with small artifacts, we train a single global model using the FEX for all types of frames, whereas for low-rate videos with large artifacts, different models are trained using both FEX and FEN for different types of frames. Finally, we propose an adaptive enhancing mechanism which is switchable between the CNN-based and the conventional methods. We selectively apply the CNN model to some frames or some regions in a frame. Experimental results show that the proposed scheme outperforms state-of-the-art work in coding efficiency, while the computational complexity is acceptable after GPU acceleration. [ABSTRACT FROM AUTHOR]

In this work, we propose a Wavelet-based Deep Auto Encoder-Decoder Network (WDAED) based image compression which takes care of the various frequency components present in an image. Specifically, we demonstrate improvements over prior approaches utilizing this framework by introducing: (a) wavelet transform pre-processing for decomposing image into different frequencies for their separate processing (b) a very deep super-resolution network as a decoder of the convolutional autoencoder in order to achieve a good quality decompressed image. The end-to-end learning is performed for four wavelet sub-bands in parallel, minimizing the computational time. The encoder compresses the image by generating the latent space representations, whereas the decoder transforms the latent space to image space. The algorithm has been tested on various standard datasets i.e., ImageNet, Set 5, Set 14, Live 1, Kodak, Classic 5, General 100 and CLIC 2019 dataset. The proposed algorithm clearly exhibited the compression performance improvement of approximately 5%, 5.5%, and 13% in terms of PSNR, PSNRB and SSIM respectively. [ABSTRACT FROM AUTHOR]