arXiv subjects

Jing Han

Publications and source records attributed to Jing Han.

At least 73 records · Page 4Linked to original sources

arXiv2020

LiDAR Data Enrichment Using Deep Learning Based on High-Resolution Image: An Approach to Achieve High-Performance LiDAR SLAM Using Low-cost LiDAR

LiDAR-based SLAM algorithms are extensively studied to providing robust and accurate positioning for autonomous driving vehicles (ADV) in the past decades. Satisfactory performance can be obtained using high-grade 3D LiDAR with 64 channels, which can provide dense point clouds. Unfortunately, the high price significantly prevents its extensive commercialization in ADV. The cost-effective 3D LiDAR with 16 channels is a promising replacement. However, only limited and sparse point clouds can be provided by the 16 channels LiDAR, which cannot guarantee sufficient positioning accuracy for ADV in challenging dynamic environments. The high-resolution image from the low-cost camera can provide ample information about the surroundings. However, the explicit depth information is not available from the image. Inspired by the complementariness of 3D LiDAR and camera, this paper proposes to make use of the high-resolution images from a camera to enrich the raw 3D point clouds from the low-cost 16 channels LiDAR based on a state-of-the-art deep learning algorithm. An ERFNet is firstly employed to segment the image with the aid of the raw sparse 3D point clouds. Meanwhile, the sparse convolutional neural network is employed to predict the dense point clouds based on raw sparse 3D point clouds. Then, the predicted dense point clouds are fused with the segmentation outputs from ERFnet using a novel multi-layer convolutional neural network to refine the predicted 3D point clouds. Finally, the enriched point clouds are employed to perform LiDAR SLAM based on the state-of-the-art normal distribution transform (NDT). We tested our approach on the re-edited KITTI datasets: (1)the sparse 3D point clouds are significantly enriched with a mean square error of 1.1m MSE. (2)the map generated from the LiDAR SLAM is denser which includes more details without significant accuracy loss.

cs.CV↗

arXiv2020

Great Chiral Fluorescence from Optical Duality Silver Nanostructures Enabled by 3D Laser Printing

Featured by prominent flexibility and fidelity in producing sophisticated stereoscopic structures transdimensionally, three-dimensional (3D) laser printing technique has vastly extended the toolkit for delivering diverse functional devices. Yet chiral nanoemitters heavily resorting to artificial structures that manifest efficient emission and tightly confined light-mater interactions simultaneously remains alluring but dauntingly challenging for this technique at this moment. In this work, we assert the chiral photoluminescence is implemented from silver nanostructures of optical duality in one go via a twofold three-dimensional laser printing scheme. Such laser printing protocol allows the highly desired duality by simultaneously producing uniformly distributed fluorescent silver nanoclusters and aggregated plasmonic silver nanoparticles to tightly confine chiral interactions at the nanoscale. A helical emitter of 550 nm-helix-diameter as fabricated has seen a record-high luminescence anisotropic factor with the absolute value up to 0.58, which is two orders of magnitude greater than fluorescent chiral silver clusters. This method holds great promise for future versatile applications in chiroptical nanodevices.

physics.optics↗

arXiv2020

Anapole mediated giant photothermal nonlinearity in nanostructured silicon

Featured with a plethora of electric and magnetic Mie resonances, high index dielectric nanostructures offer a versatile platform to concentrate light-matter interactions at the nanoscale. By integrating unique features of far-field scattering control and near-field concentration from radiationless anapole states, here, we demonstrate a giant photothermal nonlinearity in single subwavelength-sized silicon nanodisks. The nanoscale energy concentration and consequent near-field enhancements mediated by the anapole mode yield a reversible nonlinear scattering with a large modulation depth and a broad dynamic range, unveiling a record-high nonlinear index change up to 0.5 at mild incident light intensities on the order of MW/cm2. The observed photothermal nonlinearity showcases three orders of magnitude enhancement compared with that of unstructured bulk silicon, as well as nearly one order of magnitude higher than that through the radiative electric dipolar mode. Such nonlinear scattering can empower distinctive point spread functions in confocal reflectance imaging, offering the potential for far-field localization of nanostructured Si with an accuracy approaching 40 nm. Our findings shed new light on active silicon photonics based on optical anapoles.

physics.optics↗

arXiv2020

Synchronous locating and imaging behind scattering medium in a large depth based on deep learning

Scattering medium brings great difficulties to locate and image planar objects especially when the object has a large depth. In this letter, a novel learning-based method is presented to locate and image the object hidden behind a thin scattering diffuser. A multi-task network, named DINet, is constructed to predict the depth and the image of the hidden object from the captured speckle patterns. The provided experiments verify that the proposed method enables to locate the object with a depth mean error less than 0.05 mm, and image the object with an average PSNR above 24 dB, in a large depth ranging from 350 mm to 1150 mm. The constructed DINet can obtain multiple physical information via a single speckle pattern, including both the depth and image. Comparing with the traditional methods, it paves the way to the practical applications requiring large imaging depth of field behind scattering media.

eess.IV↗

arXiv2020

An Early Study on Intelligent Analysis of Speech under COVID-19: Severity, Sleep Quality, Fatigue, and Anxiety

The COVID-19 outbreak was announced as a global pandemic by the World Health Organisation in March 2020 and has affected a growing number of people in the past few weeks. In this context, advanced artificial intelligence techniques are brought to the fore in responding to fight against and reduce the impact of this global health crisis. In this study, we focus on developing some potential use-cases of intelligent speech analysis for COVID-19 diagnosed patients. In particular, by analysing speech recordings from these patients, we construct audio-only-based models to automatically categorise the health state of patients from four aspects, including the severity of illness, sleep quality, fatigue, and anxiety. For this purpose, two established acoustic feature sets and support vector machines are utilised. Our experiments show that an average accuracy of .69 obtained estimating the severity of illness, which is derived from the number of days in hospitalisation. We hope that this study can foster an extremely fast, low-cost, and convenient way to automatically detect the COVID-19 disease.

eess.AS↗

arXiv2019

Dynamic 3-D measurement based on fringe-to-fringe transformation using deep learning

Fringe projection profilometry (FPP) has become increasingly important in dynamic 3-D shape measurement. In FPP, it is necessary to retrieve the phase of the measured object before shape profiling. However, traditional phase retrieval techniques often require a large number of fringes, which may generate motion-induced error for dynamic objects. In this paper, a novel phase retrieval technique based on deep learning is proposed, which uses an end-to-end deep convolution neural network to transform a single or two fringes into the phase retrieval required fringes. When the object's surface is located in a restricted depth, the presented network only requires a single fringe as the input, which otherwise requires two fringes in an unrestricted depth. The proposed phase retrieval technique is first theoretically analyzed, and then numerically and experimentally verified on its applicability for dynamic 3-D measurement.

eess.IV↗

arXiv2019

SEWA DB: A Rich Database for Audio-Visual Emotion and Sentiment Research in the Wild

Natural human-computer interaction and audio-visual human behaviour sensing systems, which would achieve robust performance in-the-wild are more needed than ever as digital devices are increasingly becoming an indispensable part of our life. Accurately annotated real-world data are the crux in devising such systems. However, existing databases usually consider controlled settings, low demographic variability, and a single task. In this paper, we introduce the SEWA database of more than 2000 minutes of audio-visual data of 398 people coming from six cultures, 50% female, and uniformly spanning the age range of 18 to 65 years old. Subjects were recorded in two different contexts: while watching adverts and while discussing adverts in a video chat. The database includes rich annotations of the recordings in terms of facial landmarks, facial action units (FAU), various vocalisations, mirroring, and continuously valued valence, arousal, liking, agreement, and prototypic examples of (dis)liking. This database aims to be an extremely valuable resource for researchers in affective computing and automatic human sensing and is expected to push forward the research in human behaviour analysis, including cultural studies. Along with the database, we provide extensive baseline experiments for automatic FAU detection and automatic valence, arousal and (dis)liking intensity estimation.

cs.HC↗

arXiv2019

Learning-based real-time method to looking through scattering medium beyond the memory effect

Strong scattering medium brings great difficulties to optical imaging, which is also a problem in medical imaging and many other fields. Optical memory effect makes it possible to image through strong random scattering medium. However, this method also has the limitation of limited angle field-of-view (FOV), which prevents it from being applied in practice. In this paper, a kind of practical convolutional neural network called PDSNet is proposed, which effectively breaks through the limitation of optical memory effect on FOV. Experiments is conducted to prove that the scattered pattern can be reconstructed accurately in real-time by PDSNet, and it is widely applicable to retrieve complex objects of random scales and different scattering media.

eess.IV↗

arXiv2019

EmoBed: Strengthening Monomodal Emotion Recognition via Training with Crossmodal Emotion Embeddings

Despite remarkable advances in emotion recognition, they are severely restrained from either the essentially limited property of the employed single modality, or the synchronous presence of all involved multiple modalities. Motivated by this, we propose a novel crossmodal emotion embedding framework called EmoBed, which aims to leverage the knowledge from other auxiliary modalities to improve the performance of an emotion recognition system at hand. The framework generally includes two main learning components, i. e., joint multimodal training and crossmodal training. Both of them tend to explore the underlying semantic emotion information but with a shared recognition network or with a shared emotion embedding space, respectively. In doing this, the enhanced system trained with this approach can efficiently make use of the complementary information from other modalities. Nevertheless, the presence of these auxiliary modalities is not demanded during inference. To empirically investigate the effectiveness and robustness of the proposed framework, we perform extensive experiments on the two benchmark databases RECOLA and OMG-Emotion for the tasks of dimensional emotion regression and categorical emotion classification, respectively. The obtained results show that the proposed framework significantly outperforms related baselines in monomodal inference, and are also competitive or superior to the recently reported systems, which emphasises the importance of the proposed crossmodal learning for emotion recognition.

cs.LG↗

arXiv2019

High Sensitivity Snapshot Spectrometer Based on Deep Network Unmixing

In this paper, we present a convolution neural network based method to recover the light intensity distribution from the overlapped dispersive spectra instead of adding an extra light path to capture it directly for the first time. Then, we construct a single-path sub-Hadamard snapshot spectrometer based on our previous dual-path snapshot spectrometer. In the proposed single-path spectrometer, we use the reconstructed light intensity as the original light intensity and recover high signal-to-noise ratio spectra successfully. Compared with dual-path snapshot spectrometer, the network based single-path spectrometer has a more compact structure and maintains snapshot and high sensitivity. Abundant simulated and experimental results have demonstrated that the proposed method can obtain a better reconstructed signal-to-noise ratio spectrum than the dual-path sub-Hadamard spectrometer because of its higher light throughput.

eess.IV↗

arXiv2019

Non-destructive three-dimensional measurement of hand vein based on self-supervised network

At present, supervised stereo methods based on deep neural network have achieved impressive results. However, in some scenarios, accurate three-dimensional labels are inaccessible for supervised training. In this paper, a self-supervised network is proposed for binocular disparity matching (SDMNet), which computes dense disparity maps from stereo image pairs without disparity labels: In the self-supervised training, we match the stereo images densely to approximate the disparity maps and use them to warp the left and right images to estimate the right and left images; we build the loss function between estimated images and original images for self-supervised training, which adopts perceptual loss to help improve the quality of disparity maps in both detail and structure. Then, we use SDMNet to obtain disparities of hand vein. SDMNet has achieved excellent results on KITTI 2012, KITTI 2015, simulated vein dataset and real vein dataset, outperforming many state-of-the-art supervised matching methods.

cs.CV↗

arXiv2019

Low-Complexity Equalization of MIMO-OSDM

Orthogonal signal-division multiplexing (OSDM) is an attractive alternative to conventional orthogonal frequency-division multiplexing (OFDM) due to its enhanced ability in peak-to-average power ratio (PAPR) reduction. Combining OSDM with multiple-input multiple-output (MIMO) signaling has the potential to achieve high spectral and power efficiency. However, a direct channel equalization in this case incurs a cubic complexity, which may be expensive for practical use. To solve the problem, low-complexity per-vector and block equalization algorithms of MIMO-OSDM are proposed in this paper for time-invariant and time-varying channels, respectively. By exploiting the channel matrix structures, these algorithms have only a linear complexity in the transformed domain. Simulation results demonstrate their validity and the related performance comparisons.

eess.SP↗

arXiv2019

High-SNR snapshot multiplex spectrometer with sub-Hadamard-S matrix coding

We present a robust high signal-to-noise ratio (SNR) snapshot multiplex spectrometer with sub-Hadamard-S matrix coding. We demonstrated for the first time that the sub-Hadamard-S matrix coding could provide comparable SNR improvement with Hadamard-S matrix in Hadamard transform spectrometer (HTS). Normally, HTS should change the coding mask to obtain a reasonable spectrum result, causing unexpected time-consuming. An extra imaging path to collect the light intensity of the aperture is added in this paper. Both light intensity of the aperture and overlapped spectra are captured within one shot, turning Hadamard-S matrix coding into sub-Hadamard-S matrix coding. Simulations and experiments show that the proposed method could obtain comparable SNR improvement with the traditional HTS, maintaining snapshot.

eess.IV↗

arXiv2019

Snore-GANs: Improving Automatic Snore Sound Classification with Synthesized Data

One of the frontier issues that severely hamper the development of automatic snore sound classification (ASSC) associates to the lack of sufficient supervised training data. To cope with this problem, we propose a novel data augmentation approach based on semi-supervised conditional Generative Adversarial Networks (scGANs), which aims to automatically learn a mapping strategy from a random noise space to original data distribution. The proposed approach has the capability of well synthesizing 'realistic' high-dimensional data, while requiring no additional annotation process. To handle the mode collapse problem of GANs, we further introduce an ensemble strategy to enhance the diversity of the generated data. The systematic experiments conducted on a widely used Munich-Passau snore sound corpus demonstrate that the scGANs-based systems can remarkably outperform other classic data augmentation systems, and are also competitive to other recently reported systems for ASSC.

cs.LG↗

arXiv2019

Residual Pyramid Learning for Single-Shot Semantic Segmentation

Pixel-level semantic segmentation is a challenging task with a huge amount of computation, especially if the size of input is large. In the segmentation model, apart from the feature extraction, the extra decoder structure is often employed to recover spatial information. In this paper, we put forward a method for single-shot segmentation in a feature residual pyramid network (RPNet), which learns the main and residuals of segmentation by decomposing the label at different levels of residual blocks. Specifically speaking, we use the residual features to learn the edges and details, and the identity features to learn the main part of targets. At testing time, the predicted residuals are used to enhance the details of the top-level prediction. Residual learning blocks split the network into several shallow sub-networks which facilitates the training of the RPNet. We then evaluate the proposed method and compare it with recent state-of-the-art methods on CamVid and Cityscapes. The proposed single-shot segmentation based on RPNet achieves impressive results with high efficiency on pixel-level segmentation.

cs.CV↗

arXiv2018

Scaling Speech Enhancement in Unseen Environments with Noise Embeddings

We address the problem of speech enhancement generalisation to unseen environments by performing two manipulations. First, we embed an additional recording from the environment alone, and use this embedding to alter activations in the main enhancement subnetwork. Second, we scale the number of noise environments present at training time to 16,784 different environments. Experiment results show that both manipulations reduce word error rates of a pretrained speech recognition system and improve enhancement quality according to a number of performance measures. Specifically, our best model reduces the word error rate from 34.04% on noisy speech to 15.46% on the enhanced speech. Enhanced audio samples can be found in https://speechenhancement.page.link/samples.

eess.AS↗

arXiv2018

Dynamic Difficulty Awareness Training for Continuous Emotion Prediction

Time-continuous emotion prediction has become an increasingly compelling task in machine learning. Considerable efforts have been made to advance the performance of these systems. Nonetheless, the main focus has been the development of more sophisticated models and the incorporation of different expressive modalities (e. g., speech, face, and physiology). In this paper, motivated by the benefit of difficulty awareness in a human learning procedure, we propose a novel machine learning framework, namely, Dynamic Difficulty Awareness Training (DDAT), which sheds fresh light on the research -- directly exploiting the difficulties in learning to boost the machine learning process. The DDAT framework consists of two stages: information retrieval and information exploitation. In the first stage, we make use of the reconstruction error of input features or the annotation uncertainty to estimate the difficulty of learning specific information. The obtained difficulty level is then used in tandem with original features to update the model input in a second learning stage with the expectation that the model can learn to focus on high difficulty regions of the learning process. We perform extensive experiments on a benchmark database (RECOLA) to evaluate the effectiveness of the proposed framework. The experimental results show that our approach outperforms related baselines as well as other well-established time-continuous emotion prediction systems, which suggests that dynamically integrating the difficulty information for neural networks can help enhance the learning process.

cs.LG↗

arXiv2018

Adversarial Training in Affective Computing and Sentiment Analysis: Recent Advances and Perspectives

Over the past few years, adversarial training has become an extremely active research topic and has been successfully applied to various Artificial Intelligence (AI) domains. As a potentially crucial technique for the development of the next generation of emotional AI systems, we herein provide a comprehensive overview of the application of adversarial training to affective computing and sentiment analysis. Various representative adversarial training algorithms are explained and discussed accordingly, aimed at tackling diverse challenges associated with emotional AI systems. Further, we highlight a range of potential future research directions. We expect that this overview will help facilitate the development of adversarial training for affective computing and sentiment analysis in both the academic and industrial communities.

cs.CL↗