Search

To acquire high-resolution wide-swath (HRWS) imaging capacity, the displaced phase center multichannel azimuth beam (DPCMAB) technology is usually adopted in spaceborne synthetic aperture radar (SAR), while multichannel reconstruction must be carried out before imaging process due to azimuth nonuniform sampling. Up to now, almost all azimuth multichannel reconstruction algorithms have been mainly based on conventional hyperbolic range equation (CHRE), but the accuracy of the CHRE model is usually not suitable for the HRWS mode, especially for high resolution and large squint observation cases. In this study, the azimuth multichannel signal model based on the advanced hyperbolic range equation (AHRE) is established and analyzed. The major difference between multichannel signal models based on CHRE and AHRE is the additional time-varying phase error between azimuth channels. The time-varying phase error is small and can be ignored in the monostatic DPCMAB SAR system, but it must be considered and compensated in the distributed DPCMAB SAR system. In addition to the time-varying phase error, additional Doppler spectrum shift and extended Doppler bandwidth should be considered in the squint case during azimuth multichannel reconstruction. The azimuth multichannel reconstruction algorithm based on AHRE is proposed in this paper. Before multichannel reconstruction and combination, time-varying phase errors between azimuth channels were first compensated, and the range-frequency-dependent de-skewing function was derived to remove the two-dimension (2D) spectrum tilt to avoid azimuth under-sampling. Then, azimuth multichannel data were reconstructed according to the azimuth multichannel impulse response based on AHRE. Finally, the range-frequency dependent re-skewing function was introduced to recover the tilted 2D spectrum. Simulation results on both point and distributed targets validated the proposed azimuth multichannel reconstruction approach.

Exploiting the reactive voltage regulation capability of a doubly-fed induction wind generator (DFIG) and improving the reactive power response speed of wind turbines are important tasks in enhancing a power system’s voltage stability. This approach determines the reactive power distribution scheme of the two based on the regulation capacity of the stator and grid-side converter of a 1.5MW DFIG in the West Mongolia Power Grid. When the system reactive power demand surpasses the DFIG reactive power output limit, an additional active power control based on the reactive power difference is suggested. The improved genetic algorithm (IGA) is adopted to determine the additional controller parameter values. This control reduces the minimum output active power of the unit while ensuring that the reactive power output limit is increased to meet the reactive power demand of the system. Ultimately, the accuracy of the control strategy is validated through experiments.

Scene flow in 3D point clouds plays an important role in understanding dynamic environments. Although significant advances have been made by deep neural networks, the performance is far from satisfactory as only per-point translational motion is considered, neglecting the constraints of the rigid motion in local regions. To address the issue, we propose to introduce the motion consistency to force the smoothness among neighboring points. In addition, constraints on the rigidity of the local transformation are also added by sharing unique rigid motion parameters for all points within each local region. To this end, a high-order CRFs based relation module (Con-HCRFs) is deployed to explore both point-wise smoothness and region-wise rigidity. To empower the CRFs to have a discriminative unary term, we also introduce a position-aware flow estimation module to be incorporated into the Con-HCRFs. Comprehensive experiments on FlyingThings3D and KITTI show that our proposed framework (HCRF-Flow) achieves state-of-the-art performance and significantly outperforms previous approaches substantially., Comment: Accepted to CVPR2021

Due to the scarcity of annotated scene flow data, self-supervised scene flow learning in point clouds has attracted increasing attention. In the self-supervised manner, establishing correspondences between two point clouds to approximate scene flow is an effective approach. Previous methods often obtain correspondences by applying point-wise matching that only takes the distance on 3D point coordinates into account, introducing two critical issues: (1) it overlooks other discriminative measures, such as color and surface normal, which often bring fruitful clues for accurate matching; and (2) it often generates sub-par performance, as the matching is operated in an unconstrained situation, where multiple points can be ended up with the same corresponding point. To address the issues, we formulate this matching task as an optimal transport problem. The output optimal assignment matrix can be utilized to guide the generation of pseudo ground truth. In this optimal transport, we design the transport cost by considering multiple descriptors and encourage one-to-one matching by mass equality constraints. Also, constructing a graph on the points, a random walk module is introduced to encourage the local consistency of the pseudo labels. Comprehensive experiments on FlyingThings3D and KITTI show that our method achieves state-of-the-art performance among self-supervised learning methods. Our self-supervised method even performs on par with some supervised learning approaches, although we do not need any ground truth flow for training., Comment: Accepted to CVPR2021 (oral)

JSON is a very popular data format in many applications in Web and enterprise. Recently, many data analytical systems support the loading and querying JSON data. However, JSON parsing can be costly, which dominates the execution time of querying JSON data. Many previous studies focus on building efficient parsers to reduce this parsing cost, and little work has been done on how to reduce the occurrences of parsing. In this paper, we start with a study with a real production workload in Alibaba, which consists of over 3 million queries on JSON. Our study reveals significant temporal and spatial correlations among those queries, which result in massive redundant parsing operations among queries. Instead of repetitively parsing the JSON data, we propose to develop a cache system named Maxson for caching the JSON query results (the values evaluated from JSONPath) for reuse. Specifically, we develop effective machine learning-based predictor with combining LSTM (long shortterm memory) and CRF (conditional random field) to determine the JSONPaths to cache given the space budget. We have implemented Maxson on top of SparkSQL. We experimentally evaluate Maxson and show that 1) Maxson is able to eliminate the most of duplicate JSON parsing overhead, 2) Maxson improves end-to-end workload performance by 1.5–6.5×.

In this paper, we present our deep attention-based classification (DABC) network for robust single image depth prediction, in the context of the Robust Vision Challenge 2018 (ROB 2018) (http://www.robustvision.net/index.php). Unlike conventional depth prediction, our goal is to design a model that can perform well in both indoor and outdoor scenes with a single parameter set. However, robust depth prediction suffers from two challenging problems: (a) How to extract more discriminative features for different scenes (compared to a single scene)? (b) How to handle the large differences of depth ranges between indoor and outdoor datasets? To address these two problems, we first formulate depth prediction as a multi-class classification task and apply a softmax classifier to classify the depth label of each pixel. We then introduce a global pooling layer and a channel-wise attention mechanism to adaptively select the discriminative channels of features and to update the original features by assigning important channels with higher weights. Further, to reduce the influence of quantization errors, we employ a soft-weighted sum inference strategy for the final prediction. Experimental results on both indoor and outdoor datasets demonstrate the effectiveness of our method. It is worth mentioning that we won the 2-nd place in single image depth prediction entry of ROB 2018, in conjunction with IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 2018.

In this paper we study the problem of monocular relative depth perception in the wild. We introduce a simple yet effective method to automatically generate dense relative depth annotations from web stereo images, and propose a new dataset that consists of diverse images as well as corresponding dense relative depth maps. Further, an improved ranking loss is introduced to deal with imbalanced ordinal relations, enforcing the network to focus on a set of hard pairs. Experimental results demonstrate that our proposed approach not only achieves state-of-the-art accuracy of relative depth perception in the wild, but also benefits other dense per-pixel prediction tasks, e.g., metric depth estimation and semantic segmentation.

Raw scanned 3D point clouds are usually irregularly distributed due to the essential shortcomings of laser sensors, which therefore poses a great challenge for high-quality 3D surface reconstruction. This paper tackles this problem by proposing a local hierarchical clustering (LHC) method to improve the consistency of point distribution. Specifically, LHC consists of two steps: 1) adaptive octree-based decomposition of 3D space, and 2) hierarchical clustering. The former aims at reducing the computational complexity and the latter transforms the non-uniform point set into uniform one. Experimental results on real-world scanned point clouds validate the effectiveness of our method from both qualitative and quantitative aspects.

Based on JMX technology, this paper presents a new method of remote dynamic component configuration which improves the traditional method of using configuration files, and testifies its feasibility through certification programs. This method greatly improves information system flexibility through realizing dynamic configuration and safe remote component configuration and enhancing distributed management ability and security of information system greatly. Moreover, because it makes component configuration alteration take effect instantly at runtime, information system can be provided with uninterrupted running ability.