Location-aware networks will introduce innovative services and applications for modern convenience, applied ocean sciences, and public safety. In this paper, we establish a hybrid method for model-based and data-driven inference. We consider a cooperative localization (CL) scenario where the mobile agents in a wireless network aim to localize themselves by performing pairwise observations with other agents and by exchanging location information. A traditional method for distributed CL in large agent networks is belief propagation (BP) which is completely model-based and is known to suffer from providing inconsistent (overconfident) estimates. The proposed approach addresses these limitations by complementing BP with learned information provided by a graph neural network (GNN). We demonstrate numerically that our method can improve estimation accuracy and avoid overconfident beliefs, while its computational complexity remains comparable to BP. Notably, more consistent beliefs are obtained by not explicitly addressing overconfidence in the loss function used for training of the GNN.
The unified streaming and non-streaming two-pass (U2) end-to-end model for speech recognition has shown great performance in terms of streaming capability, accuracy, real-time factor (RTF), and latency. In this paper, we present U2++, an enhanced version of U2 to further improve the accuracy. The core idea of U2++ is to use the forward and the backward information of the labeling sequences at the same time at training to learn richer information, and combine the forward and backward prediction at decoding to give more accurate recognition results. We also proposed a new data augmentation method called SpecSub to help the U2++ model to be more accurate and robust. Our experiments show that, compared with U2, U2++ shows faster convergence at training, better robustness to the decoding method, as well as consistent 5\% - 8\% word error rate reduction gain over U2. On the experiment of AISHELL-1, we achieve a 4.63\% character error rate (CER) with a non-streaming setup and 5.05\% with a streaming setup with 320ms latency by U2++. To the best of our knowledge, 5.05\% is the best-published streaming result on the AISHELL-1 test set.
A number of deep learning based algorithms have been proposed to recover high-quality videos from low-quality compressed ones. Among them, some restore the missing details of each frame via exploring the spatiotemporal information of neighboring frames. However, these methods usually suffer from a narrow temporal scope, thus may miss some useful details from some frames outside the neighboring ones. In this paper, to boost artifact removal, on the one hand, we propose a Recursive Fusion (RF) module to model the temporal dependency within a long temporal range. Specifically, RF utilizes both the current reference frames and the preceding hidden state to conduct better spatiotemporal compensation. On the other hand, we design an efficient and effective Deformable Spatiotemporal Attention (DSTA) module such that the model can pay more effort on restoring the artifact-rich areas like the boundary area of a moving object. Extensive experiments show that our method outperforms the existing ones on the MFQE 2.0 dataset in terms of both fidelity and perceptual effect. Code is available at https://github.com/zhaominyiz/RFDA-PyTorch.
Mobile robots have become more and more popular in our daily life. In large-scale and crowded environments, how to navigate safely with localization precision is a critical problem. To solve this problem, we proposed a curiosity-based framework that can find an effective path with the consideration of human comfort, localization uncertainty, crowds, and the cost-to-go to the target. Three parts are involved in the proposed framework: the distance assessment module, the curiosity gain of the information-rich area, and the curiosity negative gain of crowded areas. The curiosity gain of the information-rich area was proposed to provoke the robot to approach localization referenced landmarks. To guarantee human comfort while coexisting with robots, we propose curiosity gain of the spacious area to bypass the crowd and maintain an appropriate distance between robots and humans. The evaluation is conducted in an unstructured environment. The results show that our method can find a feasible path, which can consider the localization uncertainty while simultaneously avoiding the crowded area.
Self-supervised learning has been successfully applied to pre-train video representations, which aims at efficient adaptation from pre-training domain to downstream tasks. Existing approaches merely leverage contrastive loss to learn instance-level discrimination. However, lack of category information will lead to hard-positive problem that constrains the generalization ability of this kind of methods. We find that the multi-task process of meta learning can provide a solution to this problem. In this paper, we propose a Meta-Contrastive Network (MCN), which combines the contrastive learning and meta learning, to enhance the learning ability of existing self-supervised approaches. Our method contains two training stages based on model-agnostic meta learning (MAML), each of which consists of a contrastive branch and a meta branch. Extensive evaluations demonstrate the effectiveness of our method. For two downstream tasks, i.e., video action recognition and video retrieval, MCN outperforms state-of-the-art approaches on UCF101 and HMDB51 datasets. To be more specific, with R(2+1)D backbone, MCN achieves Top-1 accuracies of 84.8% and 54.5% for video action recognition, as well as 52.5% and 23.7% for video retrieval.
Channel estimation in the RIS-aided massive multiuser multiple-input single-output (MU-MISO) wireless communication systems is challenging due to the passive feature of RIS and the large number of reflecting elements that incur high channel estimation overhead. To address this issue, we propose a novel cascaded channel estimation strategy with low pilot overhead by exploiting the sparsity and the correlation of multiuser cascaded channels in millimeter-wave massive MISO systems. Based on the fact that the phsical positions of the BS, the RIS and users may not change in several or even tens of consecutive channel coherence blocks, we first estimate the full channel state information (CSI) including all the angle and gain information in the first coherence block, and then only re-estimate the channel gains in the remaining coherence blocks with much less pilot overhead. In the first coherence block, we propose a two-phase channel estimation method, in which the cascaded channel of one typical user is estimated in Phase I based on the linear correlation among cascaded paths, while the cascaded channels of other users are estimated in Phase II by utilizing the partial CSI of the common base station (BS)-RIS channel obtained in Phase I. The total theoretical minimum pilot overhead in the first coherence block is $8J-2+(K-1)\left\lceil (8J-2)/L\right\rceil $, where $K$, $L$ and $J$ denote the numbers of users, paths in the BS-RIS channel and paths in the RIS-user channel, respectively. In each of the remaining coherence blocks, the minimum pilot overhead is $JK$. Moreover, the training phase shift matrices at the RIS are optimized to improve the estimation performance.
Although Generative Adversarial Networks (GANs) are successfully applied to diverse fields, training GANs on synthetic aperture radar (SAR) data is a challenging task mostly due to speckle noise. On the one hands, in a learning perspective of human's perception, it is natural to learn a task by using various information from multiple sources. However, in the previous GAN works on SAR target image generation, the information on target classes has only been used. Due to the backscattering characteristics of SAR image signals, the shapes and structures of SAR target images are strongly dependent on their pose angles. Nevertheless, the pose angle information has not been incorporated into such generative models for SAR target images. In this paper, we firstly propose a novel GAN-based multi-task learning (MTL) method for SAR target image generation, called PeaceGAN that uses both pose angle and target class information, which makes it possible to produce SAR target images of desired target classes at intended pose angles. For this, the PeaceGAN has two additional structures, a pose estimator and an auxiliary classifier, at the side of its discriminator to combine the pose and class information more efficiently. In addition, the PeaceGAN is jointly learned in an end-to-end manner as MTL with both pose angle and target class information, thus enhancing the diversity and quality of generated SAR target images The extensive experiments show that taking an advantage of both pose angle and target class learning by the proposed pose estimator and auxiliary classifier can help the PeaceGAN's generator effectively learn the distributions of SAR target images in the MTL framework, so that it can better generate the SAR target images more flexibly and faithfully at intended pose angles for desired target classes compared to the recent state-of-the-art methods.
Unpaired multimodal image-to-image translation is a task of translating a given image in a source domain into diverse images in the target domain, overcoming the limitation of one-to-one mapping. Existing multimodal translation models are mainly based on the disentangled representations with an image reconstruction loss. We propose two approaches to improve multimodal translation quality. First, we use a content representation from the source domain conditioned on a style representation from the target domain. Second, rather than using a typical image reconstruction loss, we design MILO (Mutual Information LOss), a new stochastically-defined loss function based on information theory. This loss function directly reflects the interpretation of latent variables as a random variable. We show that our proposed model Mutual Information with StOchastic Style Representation(MISO) achieves state-of-the-art performance through extensive experiments on various real-world datasets.
Most applications of deep learning techniques in medical imaging are supervised and require a large number of labeled data which is expensive and requires many hours of careful annotation by experts. In this paper, we propose an unsupervised, physics driven domain specific transporter framework with an attention mechanism to identify relevant key points with applications in ultrasound imaging. The proposed framework identifies key points that provide a concise geometric representation highlighting regions with high structural variation in ultrasound videos. We incorporate physics driven domain specific information as a feature probability map and use the radon transform to highlight features in specific orientations. The proposed framework has been trained on130 Lung ultrasound (LUS) videos and 113 Wrist ultrasound (WUS) videos and validated on 100 Lung ultrasound (LUS) videos and 58 Wrist ultrasound (WUS) videos acquired from multiple centers across the globe. Images from both datasets were independently assessed by experts to identify clinically relevant features such as A-lines, B-lines and pleura from LUS and radial metaphysis, radial epiphysis and carpal bones from WUS videos. The key points detected from both datasets showed high sensitivity (LUS = 99\% , WUS = 74\%) in detecting the image landmarks identified by experts. Also, on employing for classification of the given lung image into normal and abnormal classes, the proposed approach, even with no prior training, achieved an average accuracy of 97\% and an average F1-score of 95\% respectively on the task of co-classification with 3 fold cross-validation. With the purely unsupervised nature of the proposed approach, we expect the key point detection approach to increase the applicability of ultrasound in various examination performed in emergency and point of care.
Affective computing with Electroencephalogram (EEG) is a challenging task that requires cumbersome models to effectively learn the information contained in large-scale EEG signals, causing difficulties for real-time smart-device deployment. In this paper, we propose a novel knowledge distillation pipeline to distill EEG representations via capsule-based architectures for both classification and regression tasks. Our goal is to distill information from a heavy model to a lightweight model for subject-specific tasks. To this end, we first pre-train a large model (teacher network) on large number of training samples. Then, we employ the teacher network to learn the discriminative features embedded in capsules by adopting a lightweight model (student network) to mimic the teacher using the privileged knowledge. Such privileged information learned by the teacher contain similarities among capsules and are only available during the training stage of the student network. We evaluate the proposed architecture on two large-scale public EEG datasets, showing that our framework consistently enables student networks with different compression ratios to effectively learn from the teacher, even when provided with limited training samples. Lastly, our method achieves state-of-the-art results on one of the two datasets.