In millimeter-wave extremely large-scale multiple-input multiple-output (XL-MIMO) communications, fast and reliable beam alignment is essential for maximizing beamforming gain during data transmission. As XL-MIMO systems integrate increasingly larger antenna arrays, the electromagnetic propagation field transitions from the far-field to the near-field regime. This shift requires beam training methods that account for both angular and distance dimensions. Although recent advancements have been made in near-field beam training, existing approaches still face challenges, including high training overhead and limited robustness. The advent of advanced wireless systems has enabled the acquisition of user position information through various techniques, presenting opportunities to reduce training overhead in near-field scenarios. In this paper, we propose an adaptive Position-Aware Beam Training (PABT) algorithm designed for near-field XL-MIMO systems. The PABT algorithm explicitly incorporates position errors into its design to enhance robustness. Specifically, it begins with an initialization phase that uses noisy position estimates to narrow the beam search space. Following this, a multi-round iterative measurement approach is employed, where Bayes’ rule is used to update the posterior distribution of the polar-domain codebook based on the measurements obtained. This approach dynamically refines the candidate beam until a predefined probability threshold or a maximum number of iterations is reached. For the PABT proposed, we also derive a theoretical lower bound on the training overhead required to achieve a given probability threshold. This analysis provides valuable insights for selecting the key algorithm parameters in subsequent simulations. Numerical evaluations validate the proposed PABT algorithm, showing significant improvements in normalized beamforming gain and achievable spectral efficiency compared to baseline methods.

In Millimeter-wave extremely large-scale multiple-input multiple-output (XL-MIMO) communications, fast and reliable alignment of transceiver beams is essential for optimizing beamforming gain during data transmission. As XL-MIMO systems feature increasingly large antenna arrays, the electromagnetic propagation field shifts from the far field to the near field. In addition to angular domain beam training, near-field scenarios require distance domain beam training. However, many existing near-field beam training methods are essentially adaptations of far-field techniques, leading to issues such as excessive training overhead and suboptimal alignment. This paper introduces a position-aware beam training algorithm. In the initialization phase, position information is leveraged to reduce the beam search space. Subsequently, a multi-round iterative measurement strategy is employed. This approach updates the polar-domain codebook's posterior distribution using Bayes' rule based on measurements and dynamically adjusts the candidate beam set until the iteration termination condition is met. Numerical evaluations confirm the superiority of our proposed position-aware beam training algorithm, demonstrating significant enhancements in both normalized beamforming gain and achievable spectrum efficiency.

Optimizing beamforming is crucial in mitigating pronounced propagation loss and ensuring reliable communication at millimeter-wave (mmWave) frequencies. Traditional beam optimization methods rely on either precise channel estimation or extensive beam training in the mm Wave band, both of which entail substantial pilot overhead. To alleviate this overhead, we leverage the spatial congruence between sub-6 GHz (sub-6G) and mm Wave channels and propose a sub-6G information and few pilots aided beam prediction network (SPBPNet) through deep learning. Specifically, the proposed SPBPNet comprises two cascaded modules: i) the angular information extraction module, which extracts angular features from the available sub-6G channel estimate and maps them to a minimal set of narrow beam directions to be measured in the mm Wave band; and ii) the beam prediction module, which takes limited beam training along the selected directions and then fuses measurements in the mm Wave band with the sub-6G channel information to generate mmWave beam predictions. Numerical results demonstrate that SPBPNet efficiently maps sub-6G channel estimates to mmWave beams and achieves a superior balance between performance and pilot overhead compared to state-of-the-art benchmarks. Moreover, SPBPNet exhibits resilience to varying sub-6G channel estimates at different signal-to-noise ratio levels.

In skeleton-based action recognition, it has been a dominant paradigm to extract motion features with temporal convolution and model spatial correlations with graph convolution. However, it’s difficult for temporal convolution to capture long-range dependencies effectively. Meanwhile, commonly used multi-branch graph convolution leads to high complexity. In this paper, we propose TranSkeleton, a powerful Transformer framework which neatly unifies the spatial and temporal modeling of skeleton sequences. For temporal modeling, we propose a novel partition-aggregation temporal Transformer. It works with hierarchical temporal partition and aggregation, and can capture both long-range dependencies and subtle temporal structures effectively. A difference-aware aggregation approach is designed to reduce information loss during temporal aggregation. For spatial modeling, we propose a topology-aware spatial Transformer which utilizes the prior information of human body topology to facilitate spatial correlation modeling. Extensive experiments on two challenging benchmark datasets demonstrate that TranSkeleton notably outperforms the state of the arts.

Weakly-supervised temporal action localization (WTAL) is a long-standing and challenging research problem in video signal analysis. It is to localize the action segments in the video given only video-level labels. The key to this task is understanding how the diverse actions interact. In this paper, we propose W-ART, a relation Transformer to explicitly capture the relationships between action segments. We devise a new effective Transformer architecture and construct new training loss functions for WTAL. Further, we propose a dedicated query mechanism to satisfy the different feature preferences between classification and localization. Thanks to these designs, our W-ART can accurately localize the diverse actions even in weakly-supervised setting. Extensive evaluation and empirical analysis show that our method outperforms the state of the arts on two challenging benchmarks, Charades and THUMOS14.

Multi-label video classification (MLVC) is a long-standing and challenging research problem in video signal analysis. Generally, there exist many complex action labels in real-world videos and these actions are with inherent dependencies at both spatial and temporal domains. Motivated by this observation, we propose TranSTL, a spatial-temporal localization Transformer framework for MLVC task. In addition to leverage global action label co-occurrence, we also propose a novel plug-and-play Spatial Temporal Label Dependency (STLD) layer in TranSTL. STLD not only dynamically models the label co-occurrence in a video by self-attention mechanism, but also fully captures spatial-temporal label dependencies using cross-attention strategy. As a result, our TranSTL is able to explicitly and accurately grasp the diverse action labels at both spatial and temporal domains. Extensive evaluation and empirical analysis show that TranSTL achieves superior performance over the state of the arts on two challenging benchmarks, Charades and Multi-Thumos.

Satellite networks often use Performance Enhancing Proxies (PEPs) in order to overcome the inherent high latencies that are detrimental to TCP throughputs. Measurements of TCP performance over Satellite PEPs are lacking, both for actual PEP benefits and for interactions between PEPs and TCP congestion control algorithms. This paper presents results from experiments that assess the benefits of a PEP for a commercial satellite network, considering four TCP congestion control algorithms: Cubic, BBR, Hybla and PCC. Without the PEP, the four algorithms have similar steady state throughputs (about 70 Mb/s), but significant differences in start-up throughputs. In particular, the PEP dramatically improves (3x) start-up throughputs for TCP Cubic – the default congestion control algorithm used by most Internet servers. Overall, the PEP equalizes performance irrespective of the TCP congestion control algorithm chosen.

Exploiting convolutional neural networks for point cloud processing is quite challenging, due to the inherent irregular distribution and discrete shape representation of point clouds. To address these problems, many handcrafted convolution variants have sprung up in recent years. Though with elaborate design, these variants could be far from optimal in sufficiently capturing diverse shapes formed by discrete points. In this paper, we propose PointSeaConv, i.e., a novel differential convolution search paradigm on point clouds. It can work in a purely data-driven manner and thus is capable of auto-creating a group of suitable convolutions for geometric shape modeling. We also propose a joint optimization framework for simultaneous search of internal convolution and external architecture, and introduce epsilon-greedy algorithm to alleviate the effect of discretization error. As a result, PointSeaNet, a deep network that is sufficient to capture geometric shapes at both convolution level and architecture level, can be searched out for point cloud processing. Extensive experiments strongly evidence that our proposed PointSeaNet surpasses current handcrafted deep models on challenging benchmarks across multiple tasks with remarkable margins.

Skeleton-based gesture recognition is very challenging, as the high-level information in gesture is expressed by a sequence of complexly composite motions. Previous works often learn all the motions with a single model. In this paper, we propose to decouple the gesture into hand posture variations and hand movements, which are then modeled separately. For the former, the skeleton sequence is embedded into a 3D hand posture evolution volume (HPEV) to represent fine-grained posture variations. For the latter, the shifts of hand center and fingertips are arranged as a 2D hand movement map (HMM) to capture holistic movements. To learn from the two inhomogeneous representations for gesture recognition, we propose an end-to-end two-stream network. The HPEV stream integrates both spatial layout and temporal evolution information of hand postures by a dedicated 3D CNN, while the HMM stream develops an efficient 2D CNN to extract hand movement features. Eventually, the predictions of the two streams are aggregated with high efficiency. Extensive experiments on SHREC'17 Track, DHG-14/28 and FPHA datasets demonstrate that our method is competitive with the state-of-the-art.

We propose Spherical Hierarchical modeling of 3D point cloud. Inspired by Shape Context, we design a receptive field on each 3D point by placing a spherical coordinate on it. We sample points using the furthest point method and creating overlapping balls of points. We divide the space into radial, polar angular, and azimuthal angular bins on which we form a Spherical Hierarchy for each ball. We apply 1x1 CNN convolution on points to start the initial feature extraction. Repeated 3D CNN and max-pooling over the Spherical bins propagate contextual information until all the information is condensed in the center bin. Extensive experiments on five datasets strongly evidence that our method outperforms current models on various Point Cloud Learning tasks, including 2D/3D shape classification, 3D part segmentation, and 3D semantic segmentation.

3D points distribute in a continuous 3D space irregularly, thus directly adapting 2D image convolution to 3D points is not an easy job. Previous works often artificially divide the space into regular grids, yet it could be suboptimal to learn geometry. In this paper, we propose SPCNN, namely, Space Probing Convolutional Neural Network, which naturally generalizes image CNN to deal with point clouds. The key idea of SPCNN is learning to probe the 3D space in an adaptive manner. Specifically, we define a pool of learnable convolutional weights, and let each point in the local region learn to choose a suitable convolutional weight from the pool. This is achieved by constructing a geometry guided index-mapping function that implicitly establishes a correspondence between convolutional weights and some local regions in the neighborhood (Fig. 1). In this way, the index-mapping function learns to adaptively partition nearby space for local geometry pattern recognition. With this convolution as a basic operator, SPCNN, a hierarchical architecture can be developed for effective point cloud analysis. Extensive experiments on challenging benchmarks across three tasks demonstrate that SPCNN achieves the state-of-the-art or has competitive performance.

Point cloud analysis is very challenging, as the shape implied in irregular points is difficult to capture. In this paper, we propose RS-CNN, namely, Relation-Shape Convolutional Neural Network, which extends regular grid CNN to irregular configuration for point cloud analysis. The key to RS-CNN is learning from relation, i.e., the geometric topology constraint among points. Specifically, the convolutional weight for local point set is forced to learn a high-level relation expression from predefined geometric priors, between a sampled point from this point set and the others. In this way, an inductive local representation with explicit reasoning about the spatial layout of points can be obtained, which leads to much shape awareness and robustness. With this convolution as a basic operator, RS-CNN, a hierarchical architecture can be developed to achieve contextual shape-aware learning for point cloud analysis. Extensive experiments on challenging benchmarks across three tasks verify RS-CNN achieves the state of the arts.

Point cloud processing is very challenging, as the diverse shapes formed by irregular points are often indistinguishable. A thorough grasp of the elusive shape requires sufficiently contextual semantic information, yet few works devote to this. Here we propose DensePoint, a general architecture to learn densely contextual representation for point cloud processing. Technically, it extends regular grid CNN to irregular point configuration by generalizing a convolution operator, which holds the permutation invariance of points, and achieves efficient inductive learning of local patterns. Architecturally, it finds inspiration from dense connection mode, to repeatedly aggregate multi-level and multi-scale semantics in a deep hierarchy. As a result, densely contextual information along with rich semantics, can be acquired by DensePoint in an organic manner, making it highly effective. Extensive experiments on challenging benchmarks across four tasks, as well as thorough model analysis, verify DensePoint achieves the state of the arts.

Multi-label image classification is a fundamental but challenging task towards general visual understanding. Existing methods found the region-level cues (e.g., features from RoIs) can facilitate multi-label classification. Nevertheless, such methods usually require laborious object-level annotations (i.e., object labels and bounding boxes) for effective learning of the object-level visual features. In this paper, we propose a novel and efficient deep framework to boost multi-label classification by distilling knowledge from weakly-supervised detection task without bounding box annotations. Specifically, given the image-level annotations, (1) we first develop a weakly-supervised detection (WSD) model, and then (2) construct an end-to-end multi-label image classification framework augmented by a knowledge distillation module that guides the classification model by the WSD model according to the class-level predictions for the whole image and the object-level visual features for object RoIs. The WSD model is the teacher model and the classification model is the student model. After this cross-task knowledge distillation, the performance of the classification model is significantly improved and the efficiency is maintained since the WSD model can be safely discarded in the test phase. Extensive experiments on two large-scale datasets (MS-COCO and NUS-WIDE) show that our framework achieves superior performances over the state-of-the-art methods on both performance and efficiency.

Semantic labeling for the very high resolution (VHR) image of urban areas is challenging, because of many complex manmade objects with different materials and fine-structured objects located together. Under the framework of convolutional neural networks (CNNs), this paper proposes a novel end-to-end network for semantic labeling. Specifically, our network not only improves the labeling accuracy of complex manmade objects by aggregating multiple context semantics with a cascaded architecture, but also refines fine-structured objects by utilizing the low-level detail in shallow layers of CNNs with a hierarchical pyramid structure. Throughout the network, a dedicated residual correction scheme is employed to amend the latent fitting residual. As a result of these specific components, the whole model works in a global-to-local and coarse-to-fine manner. Experimental results show that our network outperforms the state-of-the-art methods on the large-scale ISPRS Vaihingen 2D Semantic Labeling Challenge dataset.