In this paper, we propose SS-NeRF, the end-to-end Neural Radiance Field (NeRF)-based architectures for high-quality physically based rendering with sparse inputs. We modify the classical spectral rendering into two main steps, 1) the generation of a series of spectrum maps spanning different wavelengths, 2) the combination of these spectrum maps for the RGB output. The proposed architecture follows these two steps through the proposed multi-layer perceptron (MLP)-based architecture (SpectralMLP) and spectrum attention UNet (SAUNet). Given the ray origin and the ray direction, the SpectralMLP constructs the spectral radiance field to obtain spectrum maps of novel views, which are then sent to the SAUNet to produce RGB images of white-light illumination. Applying NeRF to build up the spectral rendering is a more physically-based way from the perspective of ray-tracing. Further, the spectral radiance fields decompose difficult scenes and improve the performance of NeRF-based methods. Previous baseline, such as SpectralNeRF, outperforms recent methods in synthesizing novel views but requires relatively dense viewpoints for accurate scene reconstruction. To tackle this, we propose SS-NeRF to enhance the detail of scene representation with sparse inputs. In SS-NeRF, we first design the depth-aware continuity to optimize the reconstruction based on single-view depth predictions. Then, the geometric-projected consistency is introduced to optimize the multi-view geometry alignment. Additionally, we introduce a superpixel-aligned consistency to ensure that the average color within each superpixel region remains consistent. Comprehensive experimental results demonstrate that the proposed method is superior to recent state-of-the-art methods when synthesizing new views on both synthetic and real-world datasets.

Affine frequency division multiplexing (AFDM) has emerged as a promising technology for high-mobility scenarios, offering reliable performance in doubly selective channels. However, the computational complexity of the maximum likelihood (ML) detection scheme renders it impractical for real-time AFDM applications. To address this, we propose a low-complexity AFDM symbol detection algorithm based on expectation propagation (EP) in this paper. The proposed EP-based detection scheme iteratively updates messages to approximate the ML result, reducing computational complexity from exponential to cubic order. By exploiting the sparse and quasi-banded structure of the channel in the discrete affine Fourier transform (DAFT) domain and employing matrix block decomposition, lower-upper factorization, and upper triangular matrix forward substitution, we further reduce the complexity of the EP algorithm to linear order. Additionally, we optimize the EP algorithm’s performance by incorporating deep learning-based moment matching, making the algorithm more adaptive with trainable parameters for both positive and negative components. Moreover, we propose a DAFT-domain iterative detection and decoding scheme, where external information from the decoder is fed back to the detector, resulting in improved system reliability. Simulation results show that the proposed scheme achieves near-ML performance while reducing complexity by dozens of orders of magnitude compared to the ML detector, striking a balance between performance enhancement and computational complexity.

In affine frequency division multiplexing (AFDM) systems, the severe peak-to-average power ratio (PAPR) signals exist in the time domain due to the coherent superposition of numerous modulated symbols. Eventually, high PAPR signals require sophisticated and expensive power amplifiers with a very large linear range. To this end, a neural network (NN) aided intelligent transceiver optimization framework is proposed for suppressing PAPR based on the spreading AFDM structure. Specifically, the transceiver jointly optimizes the constellation geometry and associated bit labeling, the precoding NN, as well as the NN based detector. Moreover, the precoding NN is learned from a precoding approach which minimizes the variance of the instantaneous power of output signals at the transmitter. The joint optimization framework aims to achieve maximum PAPR reduction under the constraints of unit energy and spectral emission mask. Besides, to mitigate the potential inter-carrier interference during the offline training, a long short term memory based detector is designed within the optimization framework. Simulation results demonstrate that the conceived NN based optimization method achieves a significant enhancement on PAPR reduction compared with conventional approaches, while slightly improving the bit error ratio performance.

Affine frequency division multiplexing (AFDM) achieves full diversity but faces multiple-access challenges due to signal dispersion. To address this issue, we propose a power domain non-orthogonal multiple access AFDM (PD-NOMA-AFDM) system, which enables parallel transmission of multi-user signals on the same resource block through power-domain multiplexing. Furthermore, we design a binary-loop maximal ratio combining-message passing (BLMM)-based successive interference cancellation (SIC) scheme. Specifically, the inner loop fully leverages the sparsity of the AFDM equivalent channel to effectively eliminate inter-symbol interference and achieve reliable initial symbol estimation; the outer loop iteratively updates extrinsic information to compensate for performance degradation caused by banded-matrix approximation. We prove the convergence of the inner loop to the MMSE fixed point and the local convergence of the outer loop. Subsequently, by combining Lipschitz continuity and perturbation theory, we demonstrate the convergence of the overall BLMM detector to a neighborhood of the exact fixed point. The pairwise error probability analysis is then used to characterize its diversity gain and performance gap to maximum likelihood (ML) detection. To further narrow this gap, a graph neural network (GNN) is incorporated into the BLMM multi-user detection framework. This approach dynamically captures the multi-user interference (MUI) characteristics through node message interactions, thereby improving the accuracy of the approximatea posterioriprobability distribution. Simulation results show that the proposed BLMM-GNN achieves near-ML performance with strong robustness.

Index modulation (IM) techniques have been widely studied over the past decade for their ability to enhance spectral and energy efficiency by exploiting additional degrees of freedom (DoF) in waveforms. In this paper, we propose a novel subcarrier filtering orthogonal frequency-division multiplexing (OFDM) scheme, named waveform index modulation (WIM), to boost the spectral efficiency (SE) of OFDM systems without compromising other performance metrics. In WIM-OFDM, information is conveyed not only by the modulated constellation symbols but also by altering the subcarrier filter shapes, thereby utilizing an additional DoF in the OFDM signaling process. An SE-enhanced version, referred to as generalized WIM-OFDM (GWIM-OFDM), is also designed to further boost the index transmission rate by maximizing the DoF for filter selection on each subcarrier. Additionally, the optimization of subcarrier filter shapes is formulated, and a special case of subcarrier filter pair can be optimized by utilizing the proposed non-convex to convex scaling method. At the receiver, a low-complexity interference cancellation algorithm is proposed to eliminate the introduced inter-carrier interference caused by the non-orthogonal subcarrier shapes. Finally, to validate the proposed scheme, closed-form expressions for the achievable rates and the upper bound on the average bit error rate are derived to prove the superiority of our WIM-OFDM and GWIM-OFDM schemes theoretically. Monte Carlo simulation results corroborate the benefits of the proposed scheme, that is, the GWIM-OFDM scheme exhibits 4.7 to 6.1 dB performance gain, considering both bit error rate and peak-to-average power ratio, compared with the traditional OFDM and other IM benchmarking schemes under the same spectrum mask at different transmission rate scenarios.

Robots can acquire complex manipulation skills by learning policies from expert demonstrations, which is often known as vision-based imitation learning. Generating policies based on diffusion and flow matching models has been shown to be effective, particularly in robotic manipulation tasks. However, recursion-based approaches are inference inefficient in working from noise distributions to policy distributions, posing a challenging trade-off between efficiency and quality. This motivates us to propose FlowPolicy, a novel framework for fast policy generation based on consistency flow matching and 3D vision. Our approach refines the flow dynamics by normalizing the self-consistency of the velocity field, enabling the model to derive task execution policies in a single inference step. Specifically, FlowPolicy conditions on the observed 3D point cloud, where consistency flow matching directly defines straight-line flows from different time states to the same action space, while simultaneously constraining their velocity values, that is, we approximate the trajectories from noise to robot actions by normalizing the self-consistency of the velocity field within the action space, thus improving the inference efficiency. We validate the effectiveness of FlowPolicy in Adroit and Metaworld, demonstrating a 7× increase in inference speed while maintaining competitive average success rates compared to state-of-the-art methods.

To fully obtain the time-frequency diversity gain of the affine frequency division multiplexing (AFDM) system, detection algorithms that offer high performance and low complexity are essential. The maximum likelihood (ML) algorithm can achieve theoretically optimal performance. However, the exponential complexity limits its practical application. This paper designs an AFDM signal detection algorithm based on expectation propagation (EP). The proposed EP-based scheme achieves effective AFDM signal detection by iteratively updating messages to approximate the true a posterior distribution. In addition, this paper further reduces the complexity of the proposed EP-based algorithm by utilizing the characteristics of the discrete affine Fourier transform (DAFT) domain equivalent channel. Specifically, the sparsity and quasi-banded structure of the DAFT domain channel are first utilized for block processing. Subsequently, a low complexity matrix inversion operation is realized by combining the lower-upper (LU) factorization and the upper triangular matrix forward substitution algorithm. With typical AFDM system parameters, the proposed scheme reduces the complexity by 35.6 times compared to the traditional EP algorithm, while the performance is virtually unaffected. Simulation results show that the proposed scheme has a performance gain of up to 5 dB over the conventional algorithm.

In this paper, we propose a novel waveform index modulation orthogonal frequency-division multiplexing (WIM-OFDM) scheme to increase spectral efficiency for multicarrier systems. More specifically, the proposed WIM-OFDM scheme conveys not only the classic constellation symbols but also extra index bits by changing the subcarrier filtering shape for each symbol. As a key point, the optimization of the subcarrier filter shapes is formulated and a preliminary subcarrier filter pair is given to verify the performance of the proposed scheme. Our simulation results demonstrate that the proposed WIM-OFDM scheme exhibits superior performance in both peak-to-average power ratio and bit error ratio compared to conventional OFDM-IM and its dual-mode counterparts without increasing the out-of-band emission.

This paper proposes a novel joint transceiver optimization framework for multi-carrier (MC) waveform design. Unlike conventional orthogonal frequency division multiplexing, which employs memoryless modulation and fixed inverse discrete Fourier transform-based waveform generation, our approach utilizes neural network (NN)-based modulation with memory and NN-driven waveform generation at the transmitter. On the receiver side, a large-kernel attention-based NN replaces the traditional demodulation process, effectively mitigating large-span inter-carrier interference. This architecture provides enhanced flexibility for MC waveform optimization, allowing better adaptation to spectral emission mask constraints and maximizing the utilization of allocated spectrum resources. Additionally, it achieves significant spectral efficiency gains across diverse channel conditions, including additive white Gaussian noise (AWGN) and linear time-varying (LTV) channels with delay and Doppler spread. Numerical evaluations demonstrate significant bit error rate performance improvements, with up to 10 dB signal-to-noise ratio gain in LTV channels and approximately 6 dB gain in AWGN channels, underscoring the superiority of the proposed framework over state-of-the-art schemes.

In this paper, we propose SpectralNeRF, an end-to-end Neural Radiance Field (NeRF)-based architecture for high-quality physically based rendering from a novel spectral perspective. We modify the classical spectral rendering into two main steps, 1) the generation of a series of spectrum maps spanning different wavelengths, 2) the combination of these spectrum maps for the RGB output. Our SpectralNeRF follows these two steps through the proposed multi-layer perceptron (MLP)-based architecture (SpectralMLP) and Spectrum Attention UNet (SAUNet). Given the ray origin and the ray direction, the SpectralMLP constructs the spectral radiance field to obtain spectrum maps of novel views, which are then sent to the SAUNet to produce RGB images of white-light illumination. Applying NeRF to build up the spectral rendering is a more physically-based way from the perspective of ray-tracing. Further, the spectral radiance fields decompose difficult scenes and improve the performance of NeRF-based methods. Comprehensive experimental results demonstrate the proposed SpectralNeRF is superior to recent NeRF-based methods when synthesizing new views on synthetic and real datasets. The codes and datasets are available at https://github.com/liru0126/SpectralNeRF.

In the increasingly intricate and dynamic environments, it is difficult to achieve high-precision multi-target sensing with limited bandwidth resources in integrated sensing and communication. In response to the difficulty of low-power targets detection in millimeter-wave orthogonal frequency division mul-tiplexing system, a ResN et based estimation method is proposed in this paper, which has ability to realize high-precision multi-target range and velocity estimation with limited bandwidth. We first preprocess the received signal to obtain detailed channel in delay-Doppler domain. Then we utilize ResNet who has deep network and off-grid estimation capability to achieve high-precision estimation of a single high-power target. Third, an iterative structure is used to counteract the influence of this target on other targets, thereby achieving high-precision estimation of low-power targets without the number of targets. The simulation results depict that the proposed method leads lower estimation error compared to other state-of-the-art deep learning based methods and multiple signal classification algorithm in multi-target scenario. Moreover, the proposed method boasts lower computational complexity when compared to multiple signal classification algorithm.

We propose a Generative Adversarial Network (GAN)-based architecture for achieving high-quality physically based rendering (PBR). Conventional PBR relies heavily on ray tracing, which is computationally expensive in complicated environments. Some recent deep learning-based methods can improve efficiency but cannot deal with illumination variation well. In this paper, we propose PBR-GAN, an end-to-end GAN-based network that solves these problems while generating natural photo-realistic images. Two encoders (the shading encoder and albedo encoder) and two decoders (the image decoder and light decoder) are introduced to achieve our target. The two encoders and the image decoder constitute the generator that learns the mapping between the generated domain and the real domain. The light decoder produces light maps that pay more attention to the highlight and shadow regions. The discriminator aims to optimize the generator by distinguishing target images from the generated ones. Three novel loss items, concentrating on domain translation, overall shading preservation, and light map estimation, are proposed to optimize the photo-realistic outputs. Furthermore, a real dataset is collected to provide realistic information for training GAN architecture. Extensive experiments indicate that PBR-GAN can preserve the illumination variation and improve the image perceptual quality.

Point cloud registration is essential for many applications. However, existing real datasets require extremely tedious and costly annotations, yet may not provide accurate camera poses. For the synthetic datasets, they are mainly object-level, so the trained models may not generalize well to real scenes. We design SIRA-PCR, a new approach to 3D point cloud registration. First, we build a synthetic scene-level 3D registration dataset, specifically designed with physically-based and random strategies to arrange diverse objects. Second, we account for variations in different sensing mechanisms and layout placements, then formulate a sim-to-real adaptation framework with an adaptive re-sample module to simulate patterns in real point clouds. To our best knowledge, this is the first work that explores sim-to-real adaptation for point cloud registration. Extensive experiments show the SOTA performance of SIRA-PCR on widely-used indoor and out-door datasets. The code and dataset will be released on https://github.com/Chen-Suyi/SIRA_Pytorc.h

Data association is important in the point cloud registration. In this work, we propose to solve the partial-to-partial registration from a new perspective, by introducing multi-level feature interactions between the source and the reference clouds at the feature extraction stage, such that the registration can be realized without the attentions or explicit mask estimation for the overlapping detection as adopted previously. Specifically, we present FINet, a feature interactionbased structure with the capability to enable and strengthen the information associating between the inputs at multiple stages. To achieve this, we first split the features into two components, one for rotation and one for translation, based on the fact that they belong to different solution spaces, yielding a dual branches structure. Second, we insert several interaction modules at the feature extractor for the data association. Third, we propose a transformation sensitivity loss to obtain rotation-attentive and translation-attentive features. Experiments demonstrate that our method performs higher precision and robustness compared to the state-of-the-art traditional and learning-based methods. Code is available at https://github.com/megvii-research/FINet.

Video super resolution (VSR) aims to construct the high resolution (HR) frames from the low resolution ones. In this paper, we propose a new deep VSR network based on the flow-guided deformable alignment (FGDA) module and the sparsity-based temporal-spatial enhancement (STSE) module. More specifically, the FGDA module is designed to generate temporally-aligned features with the bidirectional propagation. Meanwhile, the STSE module is constructed to eliminate the alignment error for the features and strengthen the sparsity of them to enhance the spatial details to construct the high-quality HR result. In addition, we design a sparsity-based loss function to guarantee the germinated HR frames with sharp details. The experimental results demonstrate that our proposed method achieves superior performance compared with the existing popular methods.

In this letter, we propose a new local feature-based method for the detection of infrared small targets with the density feature map that can effectively suppress noise in the feature domain. First, we combine the local tetra pattern (LTrP) and the second-order LTrP to generate the density feature map. Second, we apply density peaks searching to the feature map to obtain candidate targets. Third, we generate two local features, i.e., the entropy and third-order moment, for each image patch whose center is a candidate target and then fuse them to employ the fused feature to find the real target. The experimental results demonstrate that our proposed method achieves better performance compared with the state-of-the-art approaches.

Faster-than-Nyquist signaling (FTNS) has already been shown to increase the communication capacity on certain channels such as additive white Gaussian noise and block flat multiple-input multiple-output (MIMO) Rayleigh fading channels. The following issues, however, remain unresolved: 1) whether FTNS enables a capacity increase in generalized MIMO Rayleigh fading channels that are selective in time, frequency, and space; and 2) how channel selectivities affect the capacity and if present, the FTN capacity gain. To address the issues, this paper firstly investigates the ergodic capacity of MIMO-FTN transmission over triply-selective fading channels. We derive a low-complexity approximate capacity formula and also show how it degenerates in other channel models, such as doubly-selective single-input single-output fading channels, which can be considered as the special cases of triply-selective fading channels. The capacity evaluation results obtained under different channel conditions show that: 1) MIMO-FTN outperforms MIMO-Nyquist in terms of capacity; 2) the FTN gains are nearly consistent, while the FTN gains obtained in the frequency-selective fading channels are slightly higher than those obtained in the flat fading channels.

The paper proposes a method to effectively fuse multi-exposure inputs and generate high-quality high dynamic range (HDR) images with unpaired datasets. Deep learning-based HDR image generation methods rely heavily on paired datasets. The ground truth images play a leading role in generating reasonable HDR images. Datasets without ground truth are hard to be applied to train deep neural networks. Recently, Generative Adversarial Networks (GAN) have demonstrated their potentials of translating images from source domain$X$to target domain$Y$in the absence of paired examples. In this paper, we propose a GAN-based network for solving such problems while generating enjoyable HDR results, named UPHDR-GAN. The proposed method relaxes the constraint of the paired dataset and learns the mapping from the LDR domain to the HDR domain. Although the pair data are missing, UPHDR-GAN can properly handle the ghosting artifacts caused by moving objects or misalignments with the help of the modified GAN loss, the improved discriminator network and the useful initialization phase. The proposed method preserves the details of important regions and improves the total image perceptual quality. Qualitative and quantitative comparisons against the representative methods demonstrate the superiority of the proposed UPHDR-GAN.

In this paper, we introduce a novel point-to-surface representation for 3D point cloud learning. Unlike the previous methods that mainly adopt voxel, mesh, or point coordinates, we propose to tackle this problem from a new perspective: learn a set of quadratic terms based static and global reference surfaces to describe 3D shapes, such that the coordinates of a 3D point (x, y, z) can be extended to quadratic terms (xy, xz, yz,$\ldots $) and transformed to the relationship between the local point and the global reference surfaces. Then, the static surfaces are changed into dynamic surfaces by adaptive contribution weighting to improve the descriptive capability. Towards this end, we propose our point-to-surface representation, a new representation for 3D point cloud learning that has not been attempted before, which can assemble local and global geometric information effectively by building connections between the point cloud and the learned reference surfaces. Given 3D points, we show how the reference surfaces are constructed, and how they are inserted into the 3D learning pipeline for different tasks. The experimental results confirm the effectiveness of our new representation, which has outperformed the state-of-the-art methods on the tasks of 3D classification and segmentation.

Accurate intensity estimation of tropical cyclone (TC) is important and challenging. Currently, no standardized method is available for estimating the TC intensity, usually in terms of the maximum surface wind speed, using satellite remote sensing data. This article proposes a multifeature distribution learning (MFDL) model that uses satellite microwave brightness temperatures to estimate the TC intensity. The MFDL model uses the brightness temperatures of Advanced Technology Microwave Sounder (ATMS) to generate the synthetic temperature anomaly fields of TCs at different pressure levels. These 3-D temperature anomaly data are trained by MFDL to establish the relationship between anomaly fields and TC intensity. MFDL treats the TC intensity estimation as a probability distribution of wind speed, instead of as a regression problem as in the conventional TC intensity estimation methods. In this study, 964 TCs occurred in North Atlantic (NA) and North Pacific (NP) from 2012 to 2019 are used for training and testing. The simulation results suggest that estimation accuracy is greatly improved using the preprocessed 3-D TC anomaly data. The MFDL network achieves the average mean absolute errors (MAEs) of 4.37 m/s for NA TCs and 4.92 m/s for NP TCs in the years 2018 and 2019. The results show that the MFDL network and ATMS temperature-based anomaly can be a promising means for TC intensity estimation in operational weather forecast.

The paper proposes a solution based on Generative Adversarial Network (GAN) for solving jigsaw puzzles. The problem assumes that an image is divided into equal square pieces, and asks to recover the image according to information provided by the pieces. Conventional jigsaw puzzle solvers often determine the relationships based on the boundaries of pieces, which ignore the important semantic information. In this paper, we propose JigsawGAN, a GAN-based auxiliary learning method for solving jigsaw puzzles with unpaired images (with no prior knowledge of the initial images). We design a multi-task pipeline that includes, (1) a classification branch to classify jigsaw permutations, and (2) a GAN branch to recover features to images in correct orders. The classification branch is constrained by the pseudo-labels generated according to the shuffled pieces. The GAN branch concentrates on the image semantic information, where the generator produces the natural images to fool the discriminator, while the discriminator distinguishes whether a given image belongs to the synthesized or the real target domain. These two branches are connected by a flow-based warp module that is applied to warp features to correct the order according to the classification results. The proposed method can solve jigsaw puzzles more efficiently by utilizing both semantic information and boundary information simultaneously. Qualitative and quantitative comparisons against several representative jigsaw puzzle solvers demonstrate the superiority of our method.

The application of multiple-input multiple-output (MIMO) over orthogonal time frequency space (OTFS) modulation is envisioned to provide high-data-rate wireless transmission in high-mobility environments. However, in these communication scenarios, the multiple-dimensional interference, which can generate from space, delay and Doppler domains, challenges the channel equalization and symbol detection at the MIMO-OTFS receiver. To tackle this issue, we propose a time-space domain channel equalizer, relying on the mathematical least squares minimum residual algorithm, to remove the channel distortion on data symbols. The proposed channel equalizer adopts a recursion method to achieve symbol estimates, which can realize fast convergence by leveraging the sparsity of MIMO-OTFS channel matrix. Instead of directly remapping the equalized OTFS symbols into data bits, we develop an enhanced data detection (EDD) scheme to iteratively demodulate the superposed multi-antenna signal. The EDD can not only realize the linear-complexity interference cancellation, but also efficiently reap the spatial and multi-path diversities of MIMO-OTFS channel. The simulations show the proposed channel equalization and EDD algorithms enable the MIMO-OTFS receiver to robustly demodulate multi-stream 256-ary quadrature amplitude modulation symbols, under a maximum velocity of 550 km/h at 5.9 GHz carrier frequency.

The state-of-the-art radio-frequency (RF) devices limit the deployment of extremely high-order quadrature amplitude modulation (QAM) formats (e.g., 16384-QAM) to meet the high-capacity demand on microwave backhaul links. This paper turns to faster-than-Nyquist (FTN) transmission using lower-order constellations and lower-cost RF devices as a solution to the demand. To realize low-complexity interference cancellation, we pre-equalize the FTN-induced inter-symbol interference at the transmitter by using Tomlinson-Harashima precoding (THP), while at the receiver suppressing the phase noise (PHN) generated by the RF local oscillators with pilot symbol assisted approaches. However, the THP may distort the pilots, which degrades the performance of PHN compensation. To resolve this problem, we propose two pilot designs that are distortion-free to precisely estimate the PHN samples. Moreover, we derive a closed-form expression of the symbol detection signal-to-noise ratio (SNR), in terms of the THP-FTN waveform parameters. With the SNR expression, a waveform optimization procedure is developed to maximize the SNR and enhance the achievable FTN capacity. The proposed scheme is validated in the simulated platform of 4096-QAM microwave link. The results demonstrate that the FTN signaling achieves the system capacity equivalent to that of the 16384-QAM Nyquist signaling with an SNR gain of 5.8 dB.

Point cloud registration is a key task in many computational fields. Previous correspondence matching based methods require the inputs to have distinctive geometric structures to fit a 3D rigid transformation according to point-wise sparse feature matches. However, the accuracy of transformation heavily relies on the quality of extracted features, which are prone to errors with respect to partiality and noise. In addition, they can not utilize the geometric knowledge of all the overlapping regions. On the other hand, previous global feature based approaches can utilize the entire point cloud for the registration, however they ignore the negative effect of non-overlapping points when aggregating global features. In this paper, we present OM-Net, a global feature based iterative network for partial-to-partial point cloud registration. We learn overlapping masks to reject non-overlapping regions, which converts the partial-to-partial registration to the registration of the same shape. Moreover, the previously used data is sampled only once from the CAD models for each object, resulting in the same point clouds for the source and reference. We propose a more practical manner of data generation where a CAD model is sampled twice for the source and reference, avoiding the previously prevalent over-fitting issue. Experimental results show that our method achieves state-of-the-art performance compared to traditional and deep learning based methods. Code is available at https://github.com/megvii-research/OMNet.

In this paper, we propose a new fast CU partition method for VVC intra coding based on the cross-block difference. This difference is measured by the gradient and the content of sub-blocks obtained from partition and is employed to guide the skipping of unnecessary horizontal and vertical partition modes. With this guidance, a fast determination of block partitions is accordingly achieved. Compared with VVC, our proposed method can save 41.64% (on average) encoding time with only 0.97% (on average) increase of BD-rate.

Orthogonal time frequency space (OTFS) has shown to be a promising modulation technology that achieves the robust wireless transmission in high-mobility environments. The high mobility incurred Doppler effect in OTFS system, is represented as a continuous and relatively large band in the Doppler frequency. It yields the equivalent channel responses (ECRs) in the system change significantly within one symbol block, posing a challenge to channel estimation (CE) or tracking. In order to tackle this issue, in this paper, a set of transform-domain basis functions is designed to span a low-dimensional subspace for modeling the OTFS channel. Then, the CE can be performed by estimating a few projection coefficients of ECRs in the developed subspace, with training pilots. According to the individual transmission characteristic of OTFS signal, we propose a corner-inserted pilot pattern, which targets the low pilot overhead and satisfactory CE performance. Moreover, an OTFS signal detector, leveraging the time-domain channel equalization, linear-complexity interference cancellation and delay-Doppler domain maximal ratio combining detection, is developed to retrieve the transmitted data symbols. The simulations show the precisely estimated ECRs enable the detector to ideally demodulate 256-ary quadrature amplitude modulation signaling, under a velocity of 550 km/h at 5.9 GHz carrier frequency.

Orthogonal time frequency space (OTFS) is a two-dimensional modulation scheme realized in the delay-Doppler domain, which targets the robust wireless transmissions in high-mobility environments. In such scenarios, OTFS signal suffers from multipath channel with continuous Doppler spread, which results in significant inter-symbol interference and inter-Doppler interference (IDI). In this article, we analyze the interference generation mechanism, and compare statistical distributions of the IDI in two typical cases, i.e., limited-Doppler-shift channel and continuous-Doppler-spread channel (CoDSC). Focusing on the OTFS signal transmission over the CoDSC, our study firstly indicates that the widespread IDI incurs a computational burden for the element-wise detector like the message passing in the state-of-the-art works. Addressing this challenge, we propose a block-wise OTFS receiver by exploiting the structure and characteristics of the OTFS transmission matrix. In the receiver, we deliberately design an iteration strategy among the least squares minimum residual based channel equalizer, reliability-based symbol detector and interference eliminator, which can realize fast convergence by leveraging the sparsity of channel matrix. The simulations demonstrate that, in the CoDSC, the proposed scheme achieves much less detection error, and meanwhile reduces the computational complexity by an order of magnitude, compared with the state-of-the-art OTFS receivers.

Accurate cloud detection using infrared (IR) data is very challenging due to the limitations and uncertainties from many aspects in the satellite IR remote sensing. This article proposes an end-to-end cloud detection method for the Cross-track IR Sounder (CrIS) using machine learning (ML) techniques. The brightness temperatures from paired CrIS channels in the longwave and midwave water vapor bands and the longwave and shortwave CO2bands are used. After obtaining the linear regression coefficients for each of the selected channel pairs, a complete set of CrIS full spectral resolution (FSR) cloud detection index (FCDI) is derived from the temperature difference between the regression and observation for each channel pair. It is shown that FCDI captures cloud location and structure well by comparing with the cloud products (CPs) from the Visible IR Imaging Radiometer Suite (VIIRS). After collocating FCDI with VIIRS CP, ML techniques such as the extreme learning machine, support vector machine, and multilayer perceptron are used to train the collocated FCDIs for cloud detection. Simulation results show that the accuracy of FCDI cloud detection is slightly above 80%. Moreover, the results encourage the use of water vapor bands in FCDI, in addition to CO2bands.

The paper proposes a solution to effectively handle salient regions for style transfer between unpaired datasets. Recently, Generative Adversarial Networks (GAN) have demonstrated their potentials of translating images from source domain X to target domain Y in the absence of paired examples. However, such a translation cannot guarantee to generate high perceptual quality results. Existing style transfer methods work well with relatively uniform content, they often fail to capture geometric or structural patterns that always belong to salient regions. Detail losses in structured regions and undesired artifacts in smooth regions are unavoidable even if each individual region is correctly transferred into the target style. In this paper, we propose SDP-GAN, a GAN-based network for solving such problems while generating enjoyable style transfer results. We introduce a saliency network, which is trained with the generator simultaneously. The saliency network has two functions: (1) providing constraints for content loss to increase punishment for salient regions, and (2) supplying saliency features to generator to produce coherent results. Moreover, two novel losses are proposed to optimize the generator and saliency networks. The proposed method preserves the details on important salient regions and improves the total image perceptual quality. Qualitative and quantitative comparisons against several leading prior methods demonstrates the superiority of our method.

Cloud detection is important in satellite remote sensing. It still remains challenging even though many algorithms and methods have been developed in the past years. Previously, we proposed a cloud detection index, FCDI for the S-NPP Cross-track Infrared Sounder (CrIS) full spectral resolution (FSR) data using three machine learning (ML) algorithms. Comparison with the cloud products from the Visible Infrared Imaging Radiometer Suite (VIIRS) showed that the cloud detection accuracy of FCDI of each ML algorithm was close and the best accuracy among them was about 80%. This study proposes improvements to the previous FCDI procedure. The pairing criteria of FCDI channels are redesigned to generate a more comprehensive list of the channel pairs among the longwave, mid-wave and shortwave IR bands. In this way, more atmospheric vertical levels can be represented by FCDI. In order to cover the annual variation of the real atmospheric conditions, CrIS FSR data from four seasons are utilized to derive, train, and validate the FCDI. Based on our previous study, the multilayer perceptron (MLP) is selected. Currently, the total accuracy of cloud detection using the FCDI has been improved to 84.5%. Comparison and discussion of the FCDIs are included.

Faster-than-Nyquist signaling (FTNS) is capable of improving the spectral efficiency (SE) of communication systems. However, for conventional single-carrier FTNS (SC-FTNS) in which only symbol interval is reduced, the increase of SE is very limited due to the presence of inter-symbol interference (ISI) introduced by the FTNS. To deal with this problem, this paper proposes a new time-frequency compressed SC-FTNS (TFC-SC-FTNS) scheme that includes the conventional FTNS as a special case, to improve the SE via two dimensions simultaneously: time dimension by stacking symbols closer; frequency dimension by precoding to make the FTN signal spectrum more compact. Further, an optimization subject to a spectral mask constraint is performed on the precoder to suppress the ISI, according to a mean-square-error criterion, but the optimization problem is non-convex. A nontrivial contribution in the new scheme is that the non-convex problem is transformed into a convex one by a change of variable and an addition of admissibility constraint. Simulation results demonstrate that the proposed scheme significantly outperforms the conventional FTNS in terms of achievable SE or, equivalently, reception performance at a given SE. Further, with larger constellations applied, the gains of the TFC-FTNS increase.

Faster-than-Nyquist signaling (FTNs) is capable of improving the signaling rate of communication system, while yielding the inter-symbol interference (ISI) complicating the receiver design. However, due to the unavoidable detection performance degradation when compression factor τ drops considerably below the Mazo limit, the achievable gain is limited. In this paper, preceding the FTN modulation, a precoding based data spreading is utilized to introduce an artificial interference, which aims to support a smaller τ that corresponds to achieving a higher capacity, at the cost of detection complexity. Further, we optimize the precoder by minimizing the mean square error (MSE) of the equalizer's output. Meanwhile, the problem is transformed and reformulated as a non-convex quadratically constrained and quadratic programming with one constraint (QCQP-1), where the consensus-alternating directions method of multipliers (ADMM) algorithm is utilized to iteratively pursue the solution. Simulation results justify the proposed scheme, where the capacity of 64-, 128-, and even 256-QAM Nyquist signaling can be achieved by precoding the 16-QAM FTNs, even without SNR loss at bit error rate (BER) of 10-5.

With the development of e-commerce and smart cities, utilizing unmanned vehicles to deliver packages has emerged as one of the most important methods to make customers receive packages efficiently and effectively. Hence, how to reasonably utilize multiple unmanned vehicles at the same time is a problem. Another main challenging issue is how to satisfy customers' personalized need. In this paper, we propose a novel context-aware multi-armed bandit-based online learning algorithm with active partition method for context space. To solve the massive injecting data flow problem, we utilize a tree-based structure expanding from top to bottom to choose different vehicles, which supports ever-increasing big metering datasets with historical and contextual information. We prove that our proposed context-aware online learning algorithm achieves sublinear regret performance. Experiment results show our proposal can enhance customers' satisfaction and reduce space cost tremendously.

The paper proposes a hybrid synthesis method for multi-exposure image fusion taken by hand-held cameras. Motions either due to the shaky cameras or caused by dynamic scenes should be compensated before any content fusion. The misalignment will cause blurring/ghosting artifacts in the fused result. The proposed method can deal with such motions and maintain the exposure information of each input effectively. In particular, the proposed method first applies optical flow for a coarse registration, which performs well with complex non-rigid motion but produces deformations at regions with missing correspondences. To correct such error registration, we segment images into superpixels and identify problematic alignments based on each superpixel, which is further aligned by PatchMatch. After that, the proposed method obtains a fully aligned image stack which facilitates a high-quality fusion that is free from blurring/ghosting artifacts. We compare our method with existing fusion algorithms on various challenging examples, including the static/dynamic, the indoor/outdoor and the daytime/nighttime scenes. Experiment results demonstrate the effectiveness and robustness.

Cloud detection and classification are important and challenging in satellite infrared remote sensing. This study proposes an end-to-end cloud detection and classification method for the Cross-track Infrared Sounder (CrIS) based on big data analysis and relevant machine learning schemes. Using the full spectral resolution (FSR) CrIS data, a set of FSR cloud detection indexes (FCDIs) is derived from the brightness temperatures of the selected CrIS LWIR-SWIR channel pairs. Linear and cubic regressions are compared and discussed when deriving FCDI. It's shown that FCDIs can capture clouds and structures well by comparing with the Visible Infrared Imaging Radiometer Suite (VIIRS) cloud cover/layer product. Selected from several supervised machine learning schemes, the extreme learning machine (ELM) is applied to train FCDI for cloud detection. Overall the ELM classification accuracy is about 80%. Results and discussion are provided.

In this paper, we design a new multiple description coding scheme for image signals based on our proposed compression-guided optimization. Firstly, we propose a compression-constrained adaptive filtering method to produce two descriptions for the source image, where the proposed filtering algorithm works not only to guarantee a high-quality side decoding but also make a high-efficient central decoding. Secondly, we design a compression-dependent deblocking algorithm based on the transform coefficients which are decoded from both descriptions to improve the performance for the cental decoding. Experimental results demonstrate that our proposed method achieves impressive performance gains when it is applied to image signals.

This paper studies the image fusion from multiple images taken by hand-held cameras with different exposures. The existing methods often generate unsatisfactory results, such as the blurring/ghosting artifacts due to the problematic handling of camera motions, dynamic contents, and inappropriate fusion of local regions (e.g., over or under exposed). They often require high quality image registration before fusion. However, the accurate alignment is hard to obtain in many scenarios, such as scenes with large depth variations and dynamic textures. Besides, high quality alignment is also time consuming. In this paper, we only enable a rough registration by a single homography and combine the inputs seamlessly to hide any possible misalignment. Specifically, we propose to use a Markov Random Filed (MRF) function for the labelling of all pixels, which assigns different labels to different aligned input images. During the labelling, we choose well-exposured regions and skip moving objects simultaneously. Then, we combine a Laplace image according to the labels and construct the fusion result by solving the Poisson equation. We present various challenging examples to demonstrate the effectiveness and practicability of our approach.

This paper proposes a novel 3D local descriptor, which seeks a good balance between the efficiency and the accuracy. We use the Local Reference Frame (LRF) to estimate a robust coordinate system to describe the local 3D shape. A novel Local Height Image (LHI) is defined by projecting the 3D points in the support region onto the tangent plane of the basis point. The Local Height Image Descriptor (LHID) is then defined by calculating the averaged projection distances. We further smooth the LHID to resist various kinds of interferences. We setup several experiments to assess the performance of our descriptor by comparison with the state-of-the-art algorithms. The experimental results demonstrate the effectiveness of the proposed method, which not only achieves the high accuracy as well as the robustness, but also possesses low complexity for the efficiency.

When oriented to the needs of the future fifth generation (5G) networks, orthogonal frequency division multiplexing (OFDM), as a modulation scheme, has some drawbacks: high out‐of‐band emission (OOBE) of power, relatively low spectrum efficiency, and poor flexibility for allocating resources. In this study, a subband superposed OFDM (OFDM) scheme, based on a variable granularity spectrum allocation, is proposed to divide the transmission channel into several subbands considering the compromises between single‐carrier and multicarrier spectrum utilisation. For accommodating the diverse 5G scenarios, appropriate signalling parameters can be independently configured among the subbands. A multistage polyphase interpolator is developed in the transmitter to reduce the implementation cost of time‐domain filter depressing the OOBE. Extremely narrow frequency guard intervals between subbands are realised by filtering to maximise the spectrum utilisation. At the receiver, a subband decision feedback and feedforward equaliser, relying on the subband's oversampling architecture, is designed to utilise the diversity gains in both the Doppler and the multipath delay domains. Simulation results indicate that the spectral efficiency, in terms of the spectrum utilisation rate, is increased up to and that the bit‐error‐rate performance is significantly improved for the subbands experiencing high‐speed mobile channels while preserving a relatively low computational complexity.

To make every user conveniently have access to his or her most interested jobs and candidates (recommendation items) in the current employment market, the recruitment networks need to meet the demand of fast and accurate recommendation. But a key challenge is that the total items may have a large quantity in the big data scenarios. And another problem is that the personalization of different users is diverse. In order to handle these challenges, this paper proposes a mining and prediction system for job and candidate recommendation with contextual online learning. It predicts a proper item by utilizing the feedback reward of previous users in the nearby context region. Besides that, we introduce a Monte-Carlo Tree Search (MCTS) method in which the similar items can be amalgamated into a cluster to reduce the computing load. Our algorithm can achieve sublinear regret and space complexity. Finally, some experiments are conducted to test our algorithm based on a large database from \emph{Work4} (the global leader in social and mobile recruiting), which can show the outstanding performance of our algorithm when compared with other existing algorithms.

Orthogonal frequency division multiplexing (OFDM) can not meet the diverse scenarios of the future fifth generation (5G) networks due to its high out-of-band emission (OOBE), relatively low spectrum efficiency, and poor flexibility. In this paper, a nonuniform subband superposed OFDM (NSS-OFDM) scheme, based on a variable granularity (VG) spectrum allocation technique, is proposed as a candidate waveform for 5G. The VG method is exploited to divide the transmission band into a certain amount of subbands, each of which is applied to a specified application scenario through configuring the signaling parameters. To reduce the computational complexity, a multistage polyphase subfiltering architecture is utilized. Additionally, with subtly designed filters, the OOBE is significantly suppressed, which can minimize the frequency guard intervals between subbands. At the receiver, bit-error- rate is investigated for the additive white Gaussian noise and frequency selective channel. Simulation results show that the spectral efficiency, in terms of spectrum utilization rate, is increased up to 98.85% when ignoring the BER performance loss.

We present a method to separate the background image and reflection from two photos that are taken in front of a transparent glass under slightly different viewpoints. In our method, the SIFT-flow between two images is first calculated and a motion hierarchy is constructed from the SIFT-flow at multiple levels of spatial smoothness. To distinguish background edges and reflection edges, we calculate a motion score for each edge pixel by its variance along the motion hierarchy. Alternatively, we make use of the so-called superpixels to group edge pixels into edge segments and calculate the motion scores by averaging over each segment. In the meantime, we also calculate an intensity score for each edge pixel by its gradient magnitude. We combine both motion and intensity scores to get a combination score. A binary labelling (for separation) can be obtained by thresholding the combination scores. The background image is finally reconstructed from the separated gradients. Compared to the existing approaches that require a sequence of images or a video clip for the separation, we only need two images, which largely improves its feasibility. Various challenging examples are tested to validate the effectiveness of our method.

Filtered orthogonal frequency division multiplexing (F-OFDM) system has a high computational complexity due to its high-order filter. In this paper, a subband superposed OFDM (SS-OFDM) scheme, utilizing a multistage polyphase subfiltering structure, is proposed to reduce the complexity. In the SS-OFDM system, the entire transmission channel is divided into several subbands and each subband employs relatively low order filtering to depress its out-of-band emission (OOBE). Moreover, all the subbands implement parallel transmission in time domain, which decreases the operating rate, meanwhile shortening the filter length. Simulation results show that the spectral efficiency (with respect to spectrum occupancy ratio) is increased up to about 99% for the LTE, DTMB, and DVB-T standards, while preserving a low implementation complexity.

Image interpolation refers to constructing a high-resolution (HR) image from a low-resolution (LR) image. Traditionally, an HR image can be produced from an observed LR image via the polynomial-based interpolation (bi-linear or bi-cubic interpolations, involving a small number of neighbors around each interpolated position). The advanced interpolation makes use of the so-called “geometric similarity” to design a set of optimal interpolation weighting coefficients. However, better geometric similarities can perhaps be found from a non-local area within the LR source image or even from other but similar images (possibly with higher resolutions). Based on this fact, we propose in this paper a non-local geometric similarity based interpolation scheme to construct HR images. In our proposed method, optimal weighting coefficients are determined by solving a regularized least squares problem which is built upon a number of dual reference patches drawn from the observed LR image and regularized by the variation of directional gradients of the image patch. Experimental results demonstrate that our proposed method offers a remarkable quality improvement, both objectively and subjectively.

In this paper, we propose a novel hierarchical generative model, named author-genre-topic model (AGTM), to perform satellite image annotation. Different from the existing author-topic model in which each author and topic are associated with the multinomial distributions over topics and words, in AGTM, each genre, author, and topic are associated with the multinomial distributions over authors, topics, and words, respectively. The bias of the distribution of the authors with respect to the topics can be rectified by incorporating the distribution of the genres with respect to the authors. Therefore, the classification accuracy of documents is improved when the information of genre is introduced. By representing the images with several visual words, the AGTM can be used for satellite image annotation. The labels of classes and scenes of the images correspond to the authors and the genres of the documents, respectively. The labels of classes and scenes of test images can be estimated, and the accuracy of satellite image annotation is improved when the information of scenes is introduced in the training images. Experimental results demonstrate the good performance of the proposed method.

This paper proposes a method to segment object with logos. In the method, we firstly locate the logos by SIFT matching. Then, the object boundary is extracted based on the logo location. Finally, we model the object prior based on the boundary, and introduce the prior into Markov random field segmentation method to segment the object. To verify the proposed method, we collect a logo dataset from the web such as Flickr and Google. The experimental results demonstrate the effectiveness of the proposed method.

In this paper, a novel method is proposed to perform saliency detection in news video. This method comprises bottom-up attention model which considers low level features to produce bottom-up saliency map and top-down attention model which utilizes high level factors to generate top-down saliency map. In bottom-up attention model, color image is represented as quaternion. Then the quaternion discrete cosine transform is used to detect static saliency in multi-scale and two color spaces. Meanwhile, the multi-scale local and global motion conspicuity maps are computed. To suppress the background motion noise, a novel histogram of average optical flow is proposed to calculate motion contrast. Then, the static saliency map and motion saliency map are fused after normalization. In top-down attention model, we explore high level factors of news video and generate the top-down saliency map based on these factors. Finally, the bottom-up and top-down saliency maps are integrated after normalization. Experiment results show that our method outperforms several state-of-the-art methods in saliency detection of news videos.

In this paper, a novel method is proposed to predict attention in image scenes by using a central stimuli sensitivity based saliency model. The proposed method is based on the general “center-surround” visual attention mechanism and the spatial frequency response of the human visual system (HVS). Following three biologically inspired principles, the saliency value is computed by two “scatter matrices” which are used to measure the similarity and distinctness within and between two classes, i.e., the center and surrounding regions, respectively. In order to detect salient objects with different size, the saliency of a pixel is estimated via the saliency support region of the pixel, which is the most salient region centered at the pixel with respect to the surrounding region. The proposed method which is compliant with human perceptual characteristics enables the prediction of human fixations. Experimental results on three eye tracking datasets verify the effectiveness of the method and show that the proposed method outperforms the state-of-the-art methods on the visual saliency detection task.

In this letter, we present a new face hallucination method based on similarity constraints to produce a high-resolution (HR) face image from an input low-resolution (LR) face image. This method is modeled as a local linear filtering process by incorporating four constraint functions at patch level. The first two constraints focus on checking if the training images are similar to the input face image. The third is defined in the HR face image, which is to impose the smoothness constraint between neighboring hallucinated patches. The final constraint computes the spatial distance to reduce the effect of patches that are far from the hallucinating patch. Experimental evaluation on a number of face images demonstrates the good performance of the proposed method on the face hallucination task.

The design of robust and efficient cosegmentation algorithms is challenging because of the variety and complexity of the objects and images. In this paper, we propose a new cosegmentation model by incorporating a color reward strategy and an active contour model. A new energy function corresponding to the curve is first generated with two considerations: the foreground similarity between the image pairs and the background consistency in each of the image pair. Furthermore, a new foreground similarity measurement based on the rewarding strategy is proposed. Then, we minimize the energy function value via a mutual procedure which uses dynamic priors to mutually evolve the curves. The proposed method is evaluated on many images from commonly used databases. The experimental results demonstrate that the proposed model can efficiently segment the common objects from the image pairs with generally lower error rate than many existing and conventional cosegmentation methods.

This paper proposes a method to segment object from the web images using logo detection. The method consists of three steps. In the first step, the logos are located from the original images by SIFT matching. Based on the logo location and the object shape model, the second step extracts the object boundary from the image. In the third step, we use the object boundary to model the object appearance, which is then used in the MRF based segmentation method to finally achieve the object segmentation. The key of our method is the object boundary extraction, which is achieved by searching a variation of the shape model that best fits the local edge of the image. Affine transform is used to consider the variations among the objects. Meanwhile, the Nelder-Mead simplex method with a simple initial rough search is used to run the boundary search. To verify the proposed method, we collect a LogoSeg dataset from the web such as Flickr and Google. The MOMI dataset is also used for the verification. The experimental results demonstrate that the proposed logo detection based segmentation method can improve the performance of the object segmentation.

In this paper, a new co-segmentation model by incorporating active contours based method and rewarding strategy is represented. We first generate co-segmentation energy function from two aspects. One is foreground similarity between image pairs. The other is background consistency in each single image. Then, we optimize the energy function through a mutual optimization approach. We verify the proposed method on the images commonly used in co-segmentation research. Experimental results demonstrate the effectiveness of our method.

In this letter, we propose a new method for the annotation of multispectral satellite images. This method performs the multispectral image annotation by incorporating a graphical model. To obtain the annotated image, first, we use a set of images with defined semantic concepts to represent the training set. Second, the images are represented by several visual words based on the color and texture features. Finally, an author-topic model is exploited to estimate probabilities of semantic classes for the regions in the test images and categorize them into the semantic concepts. Experimental evaluation on the multispectral images demonstrates the good performance of the proposed method on the multispectral image annotation.

Segmenting common objects that have variations in color, texture and shape is a challenging problem. In this paper, we propose a new model that efficiently segments common objects from multiple images. We first segment each original image into a number of local regions. Then, we construct a digraph based on local region similarities and saliency maps. Finally, we formulate the co-segmentation problem as the shortest path problem, and we use the dynamic programming method to solve the problem. The experimental results demonstrate that the proposed model can efficiently segment the common objects from a group of images with generally lower error rate than many existing and conventional co-segmentation methods.

In this paper, we propose a new method, called guided synthesis, to synthesize a face cartoon from a face photo. The guided synthesis is defined as a local linear model, which generates a cartoon image by incorporating the content of guidance images taken from the training set. Our synthesis operation is achieved based on four weight functions. The first is a photo-photo weight that aims to measure the similarity between an input photo patch and a training photo patch. The second is defined as a photo-cartoon weight, which is used to compute the likelihood by computing the similarity between a cartoon patch and an input photo patch. The third weight is defined in the synthesized photos, which is to set a smoothness constraint between neighboring synthesized patches. The final weight is designed to evaluate the similarity of a synthesized patch to an input patch based on the spatial distance. Experimental evaluation on a number of face photos demonstrates the good performance of the proposed method on the face cartoon synthesis.

Time selectivity of multipath channels severely degrades the performance of OFDM systems. In this paper, a piece-wise channel approximation is presented for improving the structure of the frequency-domain channel gain matrix. The proposed model is exploited to simplify the zero-forcing (ZF) equalizer as a parallel-filtering implementation with very low complexity. The linearized case of the parallel-filtering based equalizer is studied and applied to the DVB-H receiver design. The simulation indicates that the developed equalizer enforces considerably the immunity of the DVB-H receiver to the time selectivity with very little complexity increased.

In this paper, a novel segmentation algorithm is proposed to extract attention object (i.e., focus object) from Low depth of field image. In order to recognize the focus object, we first decompose the image into multiple segments that are described by visual words. Each visual word is computed from a filter bank to represent the high frequency components. The boosting method is then used to generate a strong classifier for each training image. Given a test image, we employ the voting algorithm to achieve the attention decision according to obtained strong classifiers. To extract focus objects from the test image, two-level segmentation method is proposed, which includes region and pixel levels segmentation. Experimental evaluation on test images shows that the proposed method is capable of segmenting the attention object quite effectively.