arXiv subjects

Jie Zhou

Publications and source records attributed to Jie Zhou.

At least 829 records · Page 46Linked to original sources

arXiv2020

Comprehensive Instructional Video Analysis: The COIN Dataset and Performance Evaluation

Thanks to the substantial and explosively inscreased instructional videos on the Internet, novices are able to acquire knowledge for completing various tasks. Over the past decade, growing efforts have been devoted to investigating the problem on instructional video analysis. However, the most existing datasets in this area have limitations in diversity and scale, which makes them far from many real-world applications where more diverse activities occur. To address this, we present a large-scale dataset named as "COIN" for COmprehensive INstructional video analysis. Organized with a hierarchical structure, the COIN dataset contains 11,827 videos of 180 tasks in 12 domains (e.g., vehicles, gadgets, etc.) related to our daily life. With a new developed toolbox, all the videos are annotated efficiently with a series of step labels and the corresponding temporal boundaries. In order to provide a benchmark for instructional video analysis, we evaluate plenty of approaches on the COIN dataset under five different settings. Furthermore, we exploit two important characteristics (i.e., task-consistency and ordering-dependency) for localizing important steps in instructional videos. Accordingly, we propose two simple yet effective methods, which can be easily plugged into conventional proposal-based action detection models. We believe the introduction of the COIN dataset will promote the future in-depth research on instructional video analysis for the community. Our dataset, annotation toolbox and source code are available at http://coin-dataset.github.io.

cs.CV↗

arXiv2020

Document Sub-structure in Neural Machine Translation

Current approaches to machine translation (MT) either translate sentences in isolation, disregarding the context they appear in, or model context at the level of the full document, without a notion of any internal structure the document may have. In this work we consider the fact that documents are rarely homogeneous blocks of text, but rather consist of parts covering different topics. Some documents, such as biographies and encyclopedia entries, have highly predictable, regular structures in which sections are characterised by different topics. We draw inspiration from Louis and Webber (2014) who use this information to improve statistical MT and transfer their proposal into the framework of neural MT. We compare two different methods of including information about the topic of the section within which each sentence is found: one using side constraints and the other using a cache-based model. We create and release the data on which we run our experiments - parallel corpora for three language pairs (Chinese-English, French-English, Bulgarian-English) from Wikipedia biographies, which we extract automatically, preserving the boundaries of sections within the articles.

cs.CL↗

arXiv2020

BiDet: An Efficient Binarized Object Detector

In this paper, we propose a binarized neural network learning method called BiDet for efficient object detection. Conventional network binarization methods directly quantize the weights and activations in one-stage or two-stage detectors with constrained representational capacity, so that the information redundancy in the networks causes numerous false positives and degrades the performance significantly. On the contrary, our BiDet fully utilizes the representational capacity of the binary neural networks for object detection by redundancy removal, through which the detection precision is enhanced with alleviated false positives. Specifically, we generalize the information bottleneck (IB) principle to object detection, where the amount of information in the high-level feature maps is constrained and the mutual information between the feature maps and object detection is maximized. Meanwhile, we learn sparse object priors so that the posteriors are concentrated on informative detection prediction with false positive elimination. Extensive experiments on the PASCAL VOC and COCO datasets show that our method outperforms the state-of-the-art binary neural networks by a sizable margin.

cs.CV↗

arXiv2020

Depth-Adaptive Graph Recurrent Network for Text Classification

The Sentence-State LSTM (S-LSTM) is a powerful and high efficient graph recurrent network, which views words as nodes and performs layer-wise recurrent steps between them simultaneously. Despite its successes on text representations, the S-LSTM still suffers from two drawbacks. Firstly, given a sentence, certain words are usually more ambiguous than others, and thus more computation steps need to be taken for these difficult words and vice versa. However, the S-LSTM takes fixed computation steps for all words, irrespective of their hardness. The secondary one comes from the lack of sequential information (e.g., word order) that is inherently important for natural language. In this paper, we try to address these issues and propose a depth-adaptive mechanism for the S-LSTM, which allows the model to learn how many computational steps to conduct for different words as required. In addition, we integrate an extra RNN layer to inject sequential information, which also serves as an input feature for the decision of adaptive depths. Results on the classic text classification task (24 datasets in various sizes and domains) show that our model brings significant improvements against the conventional S-LSTM and other high-performance models (e.g., the Transformer), meanwhile achieving a good accuracy-speed trade off.

cs.CL↗

arXiv2020

Open Set Modulation Recognition Based on Dual-Channel LSTM Model

Deep neural networks have achieved great success in computer vision, speech recognition and many other areas. The potential of recurrent neural networks especially the Long Short-Term Memory (LSTM) for open set communication signal modulation recognition is investigated in this letter. Time-domain sampled signals are first converted to two normalized matrices which will be fed into a four layer Dual-Channel LSTM network tailored for open set modulation recognition. With two cascaded Dual-Channel LSTM layers, the designed network can automatically learn sequence-correlated features from the raw data. With center loss and weibull distribution, proposed algorithm can recognize partial open set modulations. Experiments on the public RadioML dataset indicates that different analog and digital modulations can be effectively classified by the proposed model, while partial open set modulations can be recognized. Quantitative analysis on the dataset shows that the proposed method can achieve an average accuracy of 90.2% at varying SNR ranging from 0dB to 18dB in classifying the considered 11 classes, while accuracy of open set experiment dramatically improved by 14.2%.

eess.SP↗

arXiv2020

Joint Transmit Beamforming for Multiuser MIMO Communication and MIMO Radar

Future wireless communication systems are expected to explore spectral bands typically used by radar systems, in order to overcome spectrum congestion of traditional communication bands. Since in many applications radar and communication share the same platform, spectrum sharing can be facilitated by joint design as dual function radar-communications system. In this paper, we propose a joint transmit beamforming model for a dual-function multiple-input-multiple-output (MIMO) radar and multiuser MIMO communication transmitter sharing the spectrum and an antenna array. The proposed dual-function system transmits the weighted sum of independent radar waveform and communication symbols, forming multiple beams towards the radar targets and the communication receivers, respectively. The design of the weighting coefficients is formulated as an optimization problem whose objective is the performance of the MIMO radar transmit beamforming, while guaranteeing that the signal-to-interference-plus-noise ratio (SINR) at each communication user is higher than a given threshold. Despite the non-convexity of the proposed optimization problem, it can be relaxed into a convex one, which can be solved in polynomial time, and we prove that the relaxation is tight. Then, we propose a reduced complexity design based on zero-forcing the inter-user interference and radar interference. Unlike previous works, which focused on the transmission of communication symbols to synthesize a radar transmit beam pattern, our method provides more degrees of freedom for MIMO radar and is thus able to obtain improved radar performance, as demonstrated in our simulation study. Furthermore, the proposed dual-function scheme approaches the radar performance of the radar-only scheme, i.e., without spectrum sharing, under reasonable communication quality constraints.

eess.SP↗

arXiv2020

Bridging Text and Video: A Universal Multimodal Transformer for Video-Audio Scene-Aware Dialog

Audio-Visual Scene-Aware Dialog (AVSD) is a task to generate responses when chatting about a given video, which is organized as a track of the 8th Dialog System Technology Challenge (DSTC8). To solve the task, we propose a universal multimodal transformer and introduce the multi-task learning method to learn joint representations among different modalities as well as generate informative and fluent responses. Our method extends the natural language generation pre-trained model to multimodal dialogue generation task. Our system achieves the best performance in both objective and subjective evaluations in the challenge.

cs.CL↗

arXiv2020

SceneEncoder: Scene-Aware Semantic Segmentation of Point Clouds with A Learnable Scene Descriptor

Besides local features, global information plays an essential role in semantic segmentation, while recent works usually fail to explicitly extract the meaningful global information and make full use of it. In this paper, we propose a SceneEncoder module to impose a scene-aware guidance to enhance the effect of global information. The module predicts a scene descriptor, which learns to represent the categories of objects existing in the scene and directly guides the point-level semantic segmentation through filtering out categories not belonging to this scene. Additionally, to alleviate segmentation noise in local region, we design a region similarity loss to propagate distinguishing features to their own neighboring points with the same label, leading to the enhancement of the distinguishing ability of point-wise features. We integrate our methods into several prevailing networks and conduct extensive experiments on benchmark datasets ScanNet and ShapeNet. Results show that our methods greatly improve the performance of baselines and achieve state-of-the-art performance.

cs.CV↗

arXiv2019

P$^2$GNet: Pose-Guided Point Cloud Generating Networks for 6-DoF Object Pose Estimation

Humans are able to perform fast and accurate object pose estimation even under severe occlusion by exploiting learned object model priors from everyday life. However, most recently proposed pose estimation algorithms neglect to utilize the information of object models, often end up with limited accuracy, and tend to fall short in cluttered scenes. In this paper, we present a novel learning-based model, \underline{P}ose-Guided \underline{P}oint Cloud \underline{G}enerating Networks for 6D Object Pose Estimation (P$^2$GNet), designed to effectively exploit object model priors to facilitate 6D object pose estimation. We achieve this with an end-to-end estimation-by-generation workflow that combines the appearance information from the RGB-D image and the structure knowledge from object point cloud to enable accurate and robust pose estimation. Experiments on two commonly used benchmarks for 6D pose estimation, YCB-Video dataset and LineMOD dataset, demonstrate that P$^2$GNet outperforms the state-of-the-art method by a large margin and shows marked robustness towards heavy occlusion, while achieving real-time inference.

cs.CV↗

arXiv2019

DMRM: A Dual-channel Multi-hop Reasoning Model for Visual Dialog

Visual Dialog is a vision-language task that requires an AI agent to engage in a conversation with humans grounded in an image. It remains a challenging task since it requires the agent to fully understand a given question before making an appropriate response not only from the textual dialog history, but also from the visually-grounded information. While previous models typically leverage single-hop reasoning or single-channel reasoning to deal with this complex multimodal reasoning task, which is intuitively insufficient. In this paper, we thus propose a novel and more powerful Dual-channel Multi-hop Reasoning Model for Visual Dialog, named DMRM. DMRM synchronously captures information from the dialog history and the image to enrich the semantic representation of the question by exploiting dual-channel reasoning. Specifically, DMRM maintains a dual channel to obtain the question- and history-aware image features and the question- and image-aware dialog history features by a mulit-hop reasoning process in each channel. Additionally, we also design an effective multimodal attention to further enhance the decoder to generate more accurate responses. Experimental results on the VisDial v0.9 and v1.0 datasets demonstrate that the proposed model is effective and outperforms compared models by a significant margin.

cs.CL↗

arXiv2019

Topology of surfaces with finite Willmore energy

In this paper, we study the critical case of the Allard regularity theorem. Combining with Reifenberg's topological disk theorem, we get a critical Allard-Reifenberg type regularity theorem. As a main result, we get the topological finiteness for a class of properly immersed surfaces in $\mathbb{R}^n$ with finite Willmore energy. Especially, we prove a removability of singularity of multiplicity one surface with finite Willmore energy and a uniqueness theorem of the catenoid under no a priori topological finiteness assumption.

math.DG↗

arXiv2019

Hardness-Aware Deep Metric Learning

This paper presents a hardness-aware deep metric learning (HDML) framework. Most previous deep metric learning methods employ the hard negative mining strategy to alleviate the lack of informative samples for training. However, this mining strategy only utilizes a subset of training data, which may not be enough to characterize the global geometry of the embedding space comprehensively. To address this problem, we perform linear interpolation on embeddings to adaptively manipulate their hard levels and generate corresponding label-preserving synthetics for recycled training, so that information buried in all samples can be fully exploited and the metric is always challenged with proper difficulty. Our method achieves very competitive performance on the widely used CUB-200-2011, Cars196, and Stanford Online Products datasets.

cs.CV↗

arXiv2019

FVNet: 3D Front-View Proposal Generation for Real-Time Object Detection from Point Clouds

3D object detection from raw and sparse point clouds has been far less treated to date, compared with its 2D counterpart. In this paper, we propose a novel framework called FVNet for 3D front-view proposal generation and object detection from point clouds. It consists of two stages: generation of front-view proposals and estimation of 3D bounding box parameters. Instead of generating proposals from camera images or bird's-eye-view maps, we first project point clouds onto a cylindrical surface to generate front-view feature maps which retains rich information. We then introduce a proposal generation network to predict 3D region proposals from the generated maps and further extrude objects of interest from the whole point cloud. Finally, we present another network to extract the point-wise features from the extruded object points and regress the final 3D bounding box parameters in the canonical coordinates. Our framework achieves real-time performance with 12ms per point cloud sample. Extensive experiments on the 3D detection benchmark KITTI show that the proposed architecture outperforms state-of-the-art techniques which take either camera images or point clouds as input, in terms of accuracy and inference time.

cs.CV↗

arXiv2019

Minimizing the Bag-of-Ngrams Difference for Non-Autoregressive Neural Machine Translation

Non-Autoregressive Neural Machine Translation (NAT) achieves significant decoding speedup through generating target words independently and simultaneously. However, in the context of non-autoregressive translation, the word-level cross-entropy loss cannot model the target-side sequential dependency properly, leading to its weak correlation with the translation quality. As a result, NAT tends to generate influent translations with over-translation and under-translation errors. In this paper, we propose to train NAT to minimize the Bag-of-Ngrams (BoN) difference between the model output and the reference sentence. The bag-of-ngrams training objective is differentiable and can be efficiently calculated, which encourages NAT to capture the target-side sequential dependency and correlates well with the translation quality. We validate our approach on three translation tasks and show that our approach largely outperforms the NAT baseline by about 5.0 BLEU scores on WMT14 En$\leftrightarrow$De and about 2.5 BLEU scores on WMT16 En$\leftrightarrow$Ro.

cs.CL↗

arXiv2019

Enhancing Context Modeling with a Query-Guided Capsule Network for Document-level Translation

Context modeling is essential to generate coherent and consistent translation for Document-level Neural Machine Translations. The widely used method for document-level translation usually compresses the context information into a representation via hierarchical attention networks. However, this method neither considers the relationship between context words nor distinguishes the roles of context words. To address this problem, we propose a query-guided capsule networks to cluster context information into different perspectives from which the target translation may concern. Experiment results show that our method can significantly outperform strong baselines on multiple data sets of different domains.

cs.CL↗

arXiv2019

Measuring and Relieving the Over-smoothing Problem for Graph Neural Networks from the Topological View

Graph Neural Networks (GNNs) have achieved promising performance on a wide range of graph-based tasks. Despite their success, one severe limitation of GNNs is the over-smoothing issue (indistinguishable representations of nodes in different classes). In this work, we present a systematic and quantitative study on the over-smoothing issue of GNNs. First, we introduce two quantitative metrics, MAD and MADGap, to measure the smoothness and over-smoothness of the graph nodes representations, respectively. Then, we verify that smoothing is the nature of GNNs and the critical factor leading to over-smoothness is the low information-to-noise ratio of the message received by the nodes, which is partially determined by the graph topology. Finally, we propose two methods to alleviate the over-smoothing issue from the topological view: (1) MADReg which adds a MADGap-based regularizer to the training objective;(2) AdaGraph which optimizes the graph topology based on the model predictions. Extensive experiments on 7 widely-used graph datasets with 10 typical GNN models show that the two proposed methods are effective for relieving the over-smoothing issue, thus improving the performance of various GNN models.

cs.LG↗

arXiv2019

Multi-Zone Unit for Recurrent Neural Networks

Recurrent neural networks (RNNs) have been widely used to deal with sequence learning problems. The input-dependent transition function, which folds new observations into hidden states to sequentially construct fixed-length representations of arbitrary-length sequences, plays a critical role in RNNs. Based on single space composition, transition functions in existing RNNs often have difficulty in capturing complicated long-range dependencies. In this paper, we introduce a new Multi-zone Unit (MZU) for RNNs. The key idea is to design a transition function that is capable of modeling multiple space composition. The MZU consists of three components: zone generation, zone composition, and zone aggregation. Experimental results on multiple datasets of the character-level language modeling task and the aspect-based sentiment analysis task demonstrate the superiority of the MZU.

cs.CL↗

arXiv2019

Semantic Graph Convolutional Network for Implicit Discourse Relation Classification

Implicit discourse relation classification is of great importance for discourse parsing, but remains a challenging problem due to the absence of explicit discourse connectives communicating these relations. Modeling the semantic interactions between the two arguments of a relation has proven useful for detecting implicit discourse relations. However, most previous approaches model such semantic interactions from a shallow interactive level, which is inadequate on capturing enough semantic information. In this paper, we propose a novel and effective Semantic Graph Convolutional Network (SGCN) to enhance the modeling of inter-argument semantics on a deeper interaction level for implicit discourse relation classification. We first build an interaction graph over representations of the two arguments, and then automatically extract in-depth semantic interactive information through graph convolution. Experimental results on the English corpus PDTB and the Chinese corpus CDTB both demonstrate the superiority of our model to previous state-of-the-art systems.

cs.CL↗