arXiv subjects

Xiang Li

Publications and source records attributed to Xiang Li.

At least 1,261 records · Page 70Linked to original sources

arXiv2020

Few-shot Object Detection on Remote Sensing Images

In this paper, we deal with the problem of object detection on remote sensing images. Previous methods have developed numerous deep CNN-based methods for object detection on remote sensing images and the report remarkable achievements in detection performance and efficiency. However, current CNN-based methods mostly require a large number of annotated samples to train deep neural networks and tend to have limited generalization abilities for unseen object categories. In this paper, we introduce a few-shot learning-based method for object detection on remote sensing images where only a few annotated samples are provided for the unseen object categories. More specifically, our model contains three main components: a meta feature extractor that learns to extract feature representations from input images, a reweighting module that learn to adaptively assign different weights for each feature representation from the support images, and a bounding box prediction module that carries out object detection on the reweighted feature maps. We build our few-shot object detection model upon YOLOv3 architecture and develop a multi-scale object detection framework. Experiments on two benchmark datasets demonstrate that with only a few annotated samples our model can still achieve a satisfying detection performance on remote sensing images and the performance of our model is significantly better than the well-established baseline models.

cs.CV↗

arXiv2020

Unsupervised Learning of 3D Point Set Registration

Point cloud registration is the process of aligning a pair of point sets via searching for a geometric transformation. Recent works leverage the power of deep learning for registering a pair of point sets. However, unfortunately, deep learning models often require a large number of ground truth labels for training. Moreover, for a pair of source and target point sets, existing deep learning mechanisms require explicitly designed encoders to extract both deep spatial features from unstructured point clouds and their spatial correlation representation, which is further fed to a decoder to regress the desired geometric transformation for point set alignment. To further enhance deep learning models for point set registration, this paper proposes Deep-3DAligner, a novel unsupervised registration framework based on a newly introduced deep Spatial Correlation Representation (SCR) feature. The SCR feature describes the geometric essence of the spatial correlation between source and target point sets in an encoding-free manner. More specifically, our method starts with optimizing a randomly initialized latent SCR feature, which is then decoded to a geometric transformation (i.e., rotation and translation) to align source and target point sets. Our Deep-3DAligner jointly updates the SCR feature and weights of the transformation decoder towards the minimization of an unsupervised alignment loss. We conducted experiments on the ModelNet40 datasets to validate the performance of our unsupervised Deep-3DAligner for point set registration. The results demonstrated that, even without ground truth and any assumption of a direct correspondence between source and target point sets for training, our proposed approach achieved comparative performance compared to most recent supervised state-of-the-art approaches.

cs.CG↗

arXiv2020

Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection

One-stage detector basically formulates object detection as dense classification and localization. The classification is usually optimized by Focal Loss and the box location is commonly learned under Dirac delta distribution. A recent trend for one-stage detectors is to introduce an individual prediction branch to estimate the quality of localization, where the predicted quality facilitates the classification to improve detection performance. This paper delves into the representations of the above three fundamental elements: quality estimation, classification and localization. Two problems are discovered in existing practices, including (1) the inconsistent usage of the quality estimation and classification between training and inference and (2) the inflexible Dirac delta distribution for localization when there is ambiguity and uncertainty in complex scenes. To address the problems, we design new representations for these elements. Specifically, we merge the quality estimation into the class prediction vector to form a joint representation of localization quality and classification, and use a vector to represent arbitrary distribution of box locations. The improved representations eliminate the inconsistency risk and accurately depict the flexible distribution in real data, but contain continuous labels, which is beyond the scope of Focal Loss. We then propose Generalized Focal Loss (GFL) that generalizes Focal Loss from its discrete form to the continuous version for successful optimization. On COCO test-dev, GFL achieves 45.0\% AP using ResNet-101 backbone, surpassing state-of-the-art SAPD (43.5\%) and ATSS (43.6\%) with higher or comparable inference speed, under the same backbone and training settings. Notably, our best model can achieve a single-model single-scale AP of 48.2\%, at 10 FPS on a single 2080Ti GPU. Code and models are available at https://github.com/implus/GFocal.

cs.CV↗

arXiv2020

CAST: A Correlation-based Adaptive Spectral Clustering Algorithm on Multi-scale Data

We study the problem of applying spectral clustering to cluster multi-scale data, which is data whose clusters are of various sizes and densities. Traditional spectral clustering techniques discover clusters by processing a similarity matrix that reflects the proximity of objects. For multi-scale data, distance-based similarity is not effective because objects of a sparse cluster could be far apart while those of a dense cluster have to be sufficiently close. Following [16], we solve the problem of spectral clustering on multi-scale data by integrating the concept of objects' "reachability similarity" with a given distance-based similarity to derive an objects' coefficient matrix. We propose the algorithm CAST that applies trace Lasso to regularize the coefficient matrix. We prove that the resulting coefficient matrix has the "grouping effect" and that it exhibits "sparsity". We show that these two characteristics imply very effective spectral clustering. We evaluate CAST and 10 other clustering methods on a wide range of datasets w.r.t. various measures. Experimental results show that CAST provides excellent performance and is highly robust across test cases of multi-scale data.

cs.LG↗

arXiv2020

Modeling Discourse Structure for Document-level Neural Machine Translation

Recently, document-level neural machine translation (NMT) has become a hot topic in the community of machine translation. Despite its success, most of existing studies ignored the discourse structure information of the input document to be translated, which has shown effective in other tasks. In this paper, we propose to improve document-level NMT with the aid of discourse structure information. Our encoder is based on a hierarchical attention network (HAN). Specifically, we first parse the input document to obtain its discourse structure. Then, we introduce a Transformer-based path encoder to embed the discourse structure information of each word. Finally, we combine the discourse structure information with the word embedding before it is fed into the encoder. Experimental results on the English-to-German dataset show that our model can significantly outperform both Transformer and Transformer+HAN.

cs.CL↗

arXiv2020

Height estimation from single aerial images using a deep ordinal regression network

Understanding the 3D geometric structure of the Earth's surface has been an active research topic in photogrammetry and remote sensing community for decades, serving as an essential building block for various applications such as 3D digital city modeling, change detection, and city management. Previous researches have extensively studied the problem of height estimation from aerial images based on stereo or multi-view image matching. These methods require two or more images from different perspectives to reconstruct 3D coordinates with camera information provided. In this paper, we deal with the ambiguous and unsolved problem of height estimation from a single aerial image. Driven by the great success of deep learning, especially deep convolution neural networks (CNNs), some researches have proposed to estimate height information from a single aerial image by training a deep CNN model with large-scale annotated datasets. These methods treat height estimation as a regression problem and directly use an encoder-decoder network to regress the height values. In this paper, we proposed to divide height values into spacing-increasing intervals and transform the regression problem into an ordinal regression problem, using an ordinal loss for network training. To enable multi-scale feature extraction, we further incorporate an Atrous Spatial Pyramid Pooling (ASPP) module to extract features from multiple dilated convolution layers. After that, a post-processing technique is designed to transform the predicted height map of each patch into a seamless height map. Finally, we conduct extensive experiments on ISPRS Vaihingen and Potsdam datasets. Experimental results demonstrate significantly better performance of our method compared to the state-of-the-art methods.

cs.CV↗

arXiv2020

Detecting and Analyzing Mobility Hotspots using Surface Networks

Capabilities for collecting and storing data on mobile objects have increased dramatically over the past few decades. A persistent difficulty is summarizing large collections of mobile objects. This paper develops methods for extracting and analyzing hotspots or locations with relatively high levels of mobility activity. We use kernel density estimation (KDE) to convert a large collection of mobile objects into a smooth, continuous surface. We then develop a topological algorithm to extract critical geometric features of the surface; these include critical points (peaks, pits and passes) and critical lines (ridgelines and course-lines). We connect the peaks and corresponding ridgelines to produce a surface network that summarizes the topological structure of the surface. We apply graph theoretic indices to analytically characterize the surface and its changes over time. To illustrate our approach, we apply the techniques to taxi cab data collected in Shanghai, China. We find increases in the complexity of the hotspot spatial distribution during normal activity hours in the late morning, afternoon and evening and a spike in the connectivity of the hotspot spatial distribution in the morning as taxis concentrate on servicing travel to work. These results match with scientific and anecdotal knowledge about human activity patterns in the study area.

stat.AP↗

arXiv2020

Materials Requirements of High-Speed and Low-Power Spin-Orbit-Torque Magnetic Random-Access Memory

As spin-orbit-torque magnetic random-access memory (SOT-MRAM) is gathering great interest as the next-generation low-power and high-speed on-chip cache memory applications, it is critical to analyze the magnetic tunnel junction (MTJ) properties needed to achieve sub-ns, and ~fJ write operation when integrated with CMOS access transistors. In this paper, a 2T-1MTJ cell-level modeling framework for in-plane type Y SOT-MRAM suggests that high spin Hall conductivity and moderate SOT material sheet resistance are preferred. We benchmark write energy and speed performances of type Y SOT cells based on various SOT materials experimentally reported in the literature, including heavy metals, topological insulators and semimetals. We then carry out detailed benchmarking of SOT material Pt, beta-W, and BixSe(1-x) with different thickness and resistivity. We further discuss how our 2T-1MTJ model can be expanded to analyze other variations of SOT-MRAM, including perpendicular (type Z) and type X SOT-MRAM, two-terminal SOT-MRAM, as well as spin-transfer-torque (STT) and voltage-controlled magnetic anisotropy (VCMA)-assisted SOT-MRAM. This work will provide essential guidelines for SOT-MRAM materials, devices, and circuits research in the future.

physics.app-ph↗

arXiv2020

Convolutional Neural Network for Behavioral Modeling and Predistortion of Wideband Power Amplifiers

In this paper, we propose a novel behavior model for wideband PAs using a real-valued time-delay convolutional neural network (RVTDCNN). The input data of the model are sorted and arranged as the graph composed of the in-phase and quadrature (I/Q) components and envelope-dependent terms of current and past signals. We design a pre-designed filter using the convolutional layer to extract the basis functions required for the PA forward or reverse modeling. The generated rich basis functions are modeled using a simple fully connected layer. Because of the weight sharing characteristics of the convolutional structure, the strong memory effect does not lead to a obvious increase in the complexity of the model. Meanwhile, the extraction effect of the pre-designed filter also reduces the training complexity of the model. The experimental results show that the performance of the RVTDCNN model is almost the same as the NN models and the multilayer NN models.

eess.SP↗

arXiv2020

One-Shot Object Detection without Fine-Tuning

Deep learning has revolutionized object detection thanks to large-scale datasets, but their object categories are still arguably very limited. In this paper, we attempt to enrich such categories by addressing the one-shot object detection problem, where the number of annotated training examples for learning an unseen class is limited to one. We introduce a two-stage model consisting of a first stage Matching-FCOS network and a second stage Structure-Aware Relation Module, the combination of which integrates metric learning with an anchor-free Faster R-CNN-style detection pipeline, eventually eliminating the need to fine-tune on the support images. We also propose novel training strategies that effectively improve detection performance. Extensive quantitative and qualitative evaluations were performed and our method exceeds the state-of-the-art one-shot performance consistently on multiple datasets.

cs.CV↗

arXiv2020

ConvLab-2: An Open-Source Toolkit for Building, Evaluating, and Diagnosing Dialogue Systems

We present ConvLab-2, an open-source toolkit that enables researchers to build task-oriented dialogue systems with state-of-the-art models, perform an end-to-end evaluation, and diagnose the weakness of systems. As the successor of ConvLab (Lee et al., 2019b), ConvLab-2 inherits ConvLab's framework but integrates more powerful dialogue models and supports more datasets. Besides, we have developed an analysis tool and an interactive tool to assist researchers in diagnosing dialogue systems. The analysis tool presents rich statistics and summarizes common mistakes from simulated dialogues, which facilitates error analysis and system improvement. The interactive tool provides a user interface that allows developers to diagnose an assembled dialogue system by interacting with the system and modifying the output of each system component.

cs.CL↗

arXiv2020

FSS-1000: A 1000-Class Dataset for Few-Shot Segmentation

Over the past few years, we have witnessed the success of deep learning in image recognition thanks to the availability of large-scale human-annotated datasets such as PASCAL VOC, ImageNet, and COCO. Although these datasets have covered a wide range of object categories, there are still a significant number of objects that are not included. Can we perform the same task without a lot of human annotations? In this paper, we are interested in few-shot object segmentation where the number of annotated training examples are limited to 5 only. To evaluate and validate the performance of our approach, we have built a few-shot segmentation dataset, FSS-1000, which consists of 1000 object classes with pixelwise annotation of ground-truth segmentation. Unique in FSS-1000, our dataset contains significant number of objects that have never been seen or annotated in previous datasets, such as tiny daily objects, merchandise, cartoon characters, logos, etc. We build our baseline model using standard backbone networks such as VGG-16, ResNet-101, and Inception. To our surprise, we found that training our model from scratch using FSS-1000 achieves comparable and even better results than training with weights pre-trained by ImageNet which is more than 100 times larger than FSS-1000. Both our approach and dataset are simple, effective, and easily extensible to learn segmentation of new object classes given very few annotated training examples. Dataset is available at https://github.com/HKUSTCV/FSS-1000.

cs.CV↗

arXiv2020

A kilonova associated with GRB 070809

For on-axis typical short gamma-ray bursts (sGRBs), the forward shock emission is usually so bright that renders the identification of kilonovae (also known as macronovae) in the early afterglow ($t<0.5$ d) phase rather challenging. This is why previously no thermal-like kilonova component has been identified at such early time except in the off-axis dim GRB 170817A associated with GW170817. Here we report the identification of an unusual optical radiation component in GRB 070809 at $t\sim 0.47$ d, thanks plausibly to the very-weak/subdominant forward shock emission. The optical emission with a very red spectrum is well in excess of the extrapolation of the X-ray emission that is distinguished by an unusually hard spectrum, which is at odds with the forward shock afterglow prediction but can be naturally interpreted as a kilonova. Our finding supports the speculation that kilonovae are ubiquitous , and demonstrates the possibility of revealing the neutron star merger origin with the early afterglow data of some typical sGRBs that take place well beyond the sensitive radius of the advanced gravitational wave detectors and hence the opportunity of organizing dedicated follow-up observations for events of interest.

astro-ph.HE↗

arXiv2020

Airborne LiDAR Point Cloud Classification with Graph Attention Convolution Neural Network

Airborne light detection and ranging (LiDAR) plays an increasingly significant role in urban planning, topographic mapping, environmental monitoring, power line detection and other fields thanks to its capability to quickly acquire large-scale and high-precision ground information. To achieve point cloud classification, previous studies proposed point cloud deep learning models that can directly process raw point clouds based on PointNet-like architectures. And some recent works proposed graph convolution neural network based on the inherent topology of point clouds. However, the above point cloud deep learning models only pay attention to exploring local geometric structures, yet ignore global contextual relationships among all points. In this paper, we present a graph attention convolution neural network (GACNN) that can be directly applied to the classification of unstructured 3D point clouds obtained by airborne LiDAR. Specifically, we first introduce a graph attention convolution module that incorporates global contextual information and local structural features. Based on the proposed graph attention convolution module, we further design an end-to-end encoder-decoder network, named GACNN, to capture multiscale features of the point clouds and therefore enable more accurate airborne point cloud classification. Experiments on the ISPRS 3D labeling dataset show that the proposed model achieves a new state-of-the-art performance in terms of average F1 score (71.5\%) and a satisfying overall accuracy (83.2\%). Additionally, experiments further conducted on the 2019 Data Fusion Contest Dataset by comparing with other prevalent point cloud deep learning models demonstrate the favorable generalization capability of the proposed model.

cs.CV↗

arXiv2020

Spin-Orbit-Torque Field-Effect Transistor (SOTFET): Proposal for a New Magnetoelectric Memory

Spin-based memories are attractive for their non-volatility and high durability but provide modest resistance changes, whereas semiconductor logic transistors are capable of large resistance changes, but lack memory function with high durability. The recent availability of multiferroic materials provides an opportunity to directly couple the change in spin states of a magnetic memory to a charge change in a semiconductor transistor. In this work, we propose and analyze the spin-orbit torque field-effect transistor (SOTFET), a device with the potential to significantly boost the energy efficiency of spin-based memories, and to simultaneously offer a palette of new functionalities.

physics.app-ph↗

arXiv2020

Time-constrained Adaptive Influence Maximization

The well-known influence maximization problem aims at maximizing the influence of one information cascade in a social network by selecting appropriate seed users prior to the diffusion process. In its adaptive version, additional seed users can be selected after observing certain diffusion results. On the other hand, social computing tasks are often time-critical, and therefore only the influence resulted in the early period is worthwhile, which can be naturally modeled by enforcing a time constraint. In this paper, we present an analysis of the time-constrained adaptive influence maximization problem. We show that the new problem is combinatorially different from the existing problems, and the current techniques such as submodular maximization and adaptive submodularity are unfortunately inapplicable. On the theory side, we provide the hardness results of computing the optimal policy and a lower bound on the adaptive gap. For practical solutions, from basic to advanced, we design a series of seeding policies for achieving high efficacy and scalability. Finally, we investigate the proposed solutions through extensive simulations based on real-world datasets.

cs.SI↗

arXiv2020

Negotiation-based Human-Robot Collaboration via Augmented Reality

Effective human-robot collaboration (HRC) requires extensive communication among the human and robot teammates, because their actions can potentially produce conflicts, synergies, or both. We develop a novel augmented reality (AR) interface to bridge the communication gap between human and robot teammates. Building on our AR interface, we develop an AR-mediated, negotiation-based (ARN) framework for HRC. We have conducted experiments both in simulation and on real robots in an office environment, where multiple mobile robots work on delivery tasks. The robots could not complete the tasks on their own, but sometimes need help from their human teammate, rendering human-robot collaboration necessary. Results suggest that ARN significantly reduced the human-robot team's task completion time compared to a non-AR baseline approach.

cs.RO↗

arXiv2020

TEDL: A Text Encryption Method Based on Deep Learning

Recent years have seen an increasing emphasis on information security, and various encryption methods have been proposed. However, for symmetric encryption methods, the well-known encryption techniques still rely on the key space to guarantee security and suffer from frequent key updating. Aiming to solve those problems, this paper proposes a novel text encryption method based on deep learning called TEDL, where the secret key includes hyperparameters in deep learning model and the core step of encryption is transforming input data into weights trained under hyperparameters. Firstly, both communication parties establish a word vector table by training a deep learning model according to specified hyperparameters. Then, a self-update codebook is constructed on the word vector table with the SHA-256 function and other tricks. When communication starts, encryption and decryption are equivalent to indexing and inverted indexing on the codebook, respectively, thus achieving the transformation between plaintext and ciphertext. Results of experiments and relevant analyses show that TEDL performs well for security, efficiency, generality, and has a lower demand for the frequency of key redistribution. Especially, as a supplement to current encryption methods, the time-consuming process of constructing a codebook increases the difficulty of brute-force attacks while not degrade the communication efficiency.

cs.CR↗