arXiv subjects

Yue Lu

Publications and source records attributed to Yue Lu.

51 records · Page 3Linked to original sources

arXiv2022

Context Recovery and Knowledge Retrieval: A Novel Two-Stream Framework for Video Anomaly Detection

Video anomaly detection aims to find the events in a video that do not conform to the expected behavior. The prevalent methods mainly detect anomalies by snippet reconstruction or future frame prediction error. However, the error is highly dependent on the local context of the current snippet and lacks the understanding of normality. To address this issue, we propose to detect anomalous events not only by the local context, but also according to the consistency between the testing event and the knowledge about normality from the training data. Concretely, we propose a novel two-stream framework based on context recovery and knowledge retrieval, where the two streams can complement each other. For the context recovery stream, we propose a spatiotemporal U-Net which can fully utilize the motion information to predict the future frame. Furthermore, we propose a maximum local error mechanism to alleviate the problem of large recovery errors caused by complex foreground objects. For the knowledge retrieval stream, we propose an improved learnable locality-sensitive hashing, which optimizes hash functions via a Siamese network and a mutual difference loss. The knowledge about normality is encoded and stored in hash tables, and the distance between the testing event and the knowledge representation is used to reveal the probability of anomaly. Finally, we fuse the anomaly scores from the two streams to detect anomalies. Extensive experiments demonstrate the effectiveness and complementarity of the two streams, whereby the proposed two-stream framework achieves state-of-the-art performance on four datasets.

cs.CV↗

arXiv2022

UC-OWOD: Unknown-Classified Open World Object Detection

Open World Object Detection (OWOD) is a challenging computer vision problem that requires detecting unknown objects and gradually learning the identified unknown classes. However, it cannot distinguish unknown instances as multiple unknown classes. In this work, we propose a novel OWOD problem called Unknown-Classified Open World Object Detection (UC-OWOD). UC-OWOD aims to detect unknown instances and classify them into different unknown classes. Besides, we formulate the problem and devise a two-stage object detector to solve UC-OWOD. First, unknown label-aware proposal and unknown-discriminative classification head are used to detect known and unknown objects. Then, similarity-based unknown classification and unknown clustering refinement modules are constructed to distinguish multiple unknown classes. Moreover, two novel evaluation protocols are designed to evaluate unknown-class detection. Abundant experiments and visualizations prove the effectiveness of the proposed method. Code is available at https://github.com/JohnWuzh/UC-OWOD.

cs.CV↗

arXiv2022

SGBANet: Semantic GAN and Balanced Attention Network for Arbitrarily Oriented Scene Text Recognition

Scene text recognition is a challenging task due to the complex backgrounds and diverse variations of text instances. In this paper, we propose a novel Semantic GAN and Balanced Attention Network (SGBANet) to recognize the texts in scene images. The proposed method first generates the simple semantic feature using Semantic GAN and then recognizes the scene text with the Balanced Attention Module. The Semantic GAN aims to align the semantic feature distribution between the support domain and target domain. Different from the conventional image-to-image translation methods that perform at the image level, the Semantic GAN performs the generation and discrimination on the semantic level with the Semantic Generator Module (SGM) and Semantic Discriminator Module (SDM). For target images (scene text images), the Semantic Generator Module generates simple semantic features that share the same feature distribution with support images (clear text images). The Semantic Discriminator Module is used to distinguish the semantic features between the support domain and target domain. In addition, a Balanced Attention Module is designed to alleviate the problem of attention drift. The Balanced Attention Module first learns a balancing parameter based on the visual glimpse vector and semantic glimpse vector, and then performs the balancing operation for obtaining a balanced glimpse vector. Experiments on six benchmarks, including regular datasets, i.e., IIIT5K, SVT, ICDAR2013, and irregular datasets, i.e., ICDAR2015, SVTP, CUTE80, validate the effectiveness of our proposed method.

cs.CV↗

arXiv2022

A Deep Learning Approach for Predicting Two-dimensional Soil Consolidation Using Physics-Informed Neural Networks (PINN)

Soil consolidation is closely related to seepage, stability, and settlement of geotechnical buildings and foundations, and directly affects the use and safety of superstructures. Nowadays, the unidirectional consolidation theory of soils is widely used in certain conditions and approximate calculations. The multi-directional theory of soil consolidation is more reasonable than the unidirectional theory in practical applications, but it is much more complicated in terms of index determination and solution. To address the above problem, in this paper, we propose a deep learning method using physics-informed neural networks (PINN) to predict the excess pore water pressure of two-dimensional soil consolidation. In the proposed method, (1) a fully connected neural network is constructed, (2) the computational domain, partial differential equation (PDE), and constraints are defined to generate data for model training, and (3) the PDE of two-dimensional soil consolidation and the model of the neural network is connected to reduce the loss of the model. The effectiveness of the proposed method is verified by comparison with the numerical solution of PDE for two-dimensional consolidation. Using this method, the excess pore water pressure could be predicted simply and efficiently. In addition, the method was applied to predict the soil excess pore water pressure in the foundation in a real case at Tianjin port, China. The proposed deep learning approach can be used to investigate the large and complex multi-directional soil consolidation.

cs.CE↗

arXiv2021

Learnable Locality-Sensitive Hashing for Video Anomaly Detection

Video anomaly detection (VAD) mainly refers to identifying anomalous events that have not occurred in the training set where only normal samples are available. Existing works usually formulate VAD as a reconstruction or prediction problem. However, the adaptability and scalability of these methods are limited. In this paper, we propose a novel distance-based VAD method to take advantage of all the available normal data efficiently and flexibly. In our method, the smaller the distance between a testing sample and normal samples, the higher the probability that the testing sample is normal. Specifically, we propose to use locality-sensitive hashing (LSH) to map samples whose similarity exceeds a certain threshold into the same bucket in advance. In this manner, the complexity of near neighbor search is cut down significantly. To make the samples that are semantically similar get closer and samples not similar get further apart, we propose a novel learnable version of LSH that embeds LSH into a neural network and optimizes the hash functions with contrastive learning strategy. The proposed method is robust to data imbalance and can handle the large intra-class variations in normal data flexibly. Besides, it has a good ability of scalability. Extensive experiments demonstrate the superiority of our method, which achieves new state-of-the-art results on VAD benchmarks.

cs.CV↗

arXiv2021

DG-Font: Deformable Generative Networks for Unsupervised Font Generation

Font generation is a challenging problem especially for some writing systems that consist of a large number of characters and has attracted a lot of attention in recent years. However, existing methods for font generation are often in supervised learning. They require a large number of paired data, which is labor-intensive and expensive to collect. Besides, common image-to-image translation models often define style as the set of textures and colors, which cannot be directly applied to font generation. To address these problems, we propose novel deformable generative networks for unsupervised font generation (DGFont). We introduce a feature deformation skip connection (FDSC) which predicts pairs of displacement maps and employs the predicted maps to apply deformable convolution to the low-level feature maps from the content encoder. The outputs of FDSC are fed into a mixer to generate the final results. Taking advantage of FDSC, the mixer outputs a high-quality character with a complete structure. To further improve the quality of generated images, we use three deformable convolution layers in the content encoder to learn style-invariant feature representations. Experiments demonstrate that our model generates characters in higher quality than state-of-art methods. The source code is available at https://github.com/ecnuycxie/DG-Font.

cs.CV↗

arXiv2021

Efficient Spatialtemporal Context Modeling for Action Recognition

Contextual information plays an important role in action recognition. Local operations have difficulty to model the relation between two elements with a long-distance interval. However, directly modeling the contextual information between any two points brings huge cost in computation and memory, especially for action recognition, where there is an additional temporal dimension. Inspired from 2D criss-cross attention used in segmentation task, we propose a recurrent 3D criss-cross attention (RCCA-3D) module to model the dense long-range spatiotemporal contextual information in video for action recognition. The global context is factorized into sparse relation maps. We model the relationship between points in the same line along the direction of horizon, vertical and depth at each time, which forms a 3D criss-cross structure, and duplicate the same operation with recurrent mechanism to transmit the relation between points in a line to a plane finally to the whole spatiotemporal space. Compared with the non-local method, the proposed RCCA-3D module reduces the number of parameters and FLOPs by 25% and 30% for video context modeling. We evaluate the performance of RCCA-3D with two latest action recognition networks on three datasets and make a thorough analysis of the architecture, obtaining the optimal way to factorize and fuse the relation maps. Comparisons with other state-of-the-art methods demonstrate the effectiveness and efficiency of our model.

cs.CV↗

arXiv2020

Automatic Historical Feature Generation through Tree-based Method in Ads Prediction

Historical features are important in ads click-through rate (CTR) prediction, because they account for past engagements between users and ads. In this paper, we study how to efficiently construct historical features through counting features. The key challenge of such problem lies in how to automatically identify counting keys. We propose a tree-based method for counting key selection. The intuition is that a decision tree naturally provides various combinations of features, which could be used as counting key candidate. In order to select personalized counting features, we train one decision tree model per user, and the counting keys are selected across different users with a frequency-based importance measure. To validate the effectiveness of proposed solution, we conduct large scale experiments on Twitter video advertising data. In both online learning and offline training settings, the automatically identified counting features outperform the manually curated counting features.

cs.LG↗

arXiv2020

Reveal of Domain Effect: How Visual Restoration Contributes to Object Detection in Aquatic Scenes

Underwater robotic perception usually requires visual restoration and object detection, both of which have been studied for many years. Meanwhile, data domain has a huge impact on modern data-driven leaning process. However, exactly indicating domain effect, the relation between restoration and detection remains unclear. In this paper, we generally investigate the relation of quality-diverse data domain to detection performance. In the meantime, we unveil how visual restoration contributes to object detection in real-world underwater scenes. According to our analysis, five key discoveries are reported: 1) Domain quality has an ignorable effect on within-domain convolutional representation and detection accuracy; 2) low-quality domain leads to higher generalization ability in cross-domain detection; 3) low-quality domain can hardly be well learned in a domain-mixed learning process; 4) degrading recall efficiency, restoration cannot improve within-domain detection accuracy; 5) visual restoration is beneficial to detection in the wild by reducing the domain shift between training data and real-world scenes. Finally, as an illustrative example, we successfully perform underwater object detection with an aquatic robot.

cs.CV↗

arXiv2020

Bilinear Constraint based ADMM for Mixed Poisson-Gaussian Noise Removal

In this paper, we propose new operator-splitting algorithms for the total variation regularized infimal convolution (TV-IC) model [4] in order to remove mixed Poisson-Gaussian(MPG) noise. In the existing splitting algorithm for TV-IC, an inner loop by Newton method had to be adopted for one nonlinear optimization subproblem, which increased the computation cost per outer loop. By introducing a new bilinear constraint and applying the alternating direction method of multipliers (ADMM), all subproblems of the proposed algorithms named as BCA (short for Bilinear Constraint based ADMM algorithm) and BCAf(short for a variant of BCA with fully splitting form) can be very efficiently solved; especially for the proposed BCAf, they can be calculated without any inner iterations. Under mild conditions, the convergence of the proposed BCA is investigated. Numerically, compared to existing primal-dual algorithms for the TV-IC model, the proposed algorithms, with fewer tunable parameters, converge much faster and produce comparable results meanwhile.

math.OC↗

arXiv2020

FACLSTM: ConvLSTM with Focused Attention for Scene Text Recognition

Scene text recognition has recently been widely treated as a sequence-to-sequence prediction problem, where traditional fully-connected-LSTM (FC-LSTM) has played a critical role. Due to the limitation of FC-LSTM, existing methods have to convert 2-D feature maps into 1-D sequential feature vectors, resulting in severe damages of the valuable spatial and structural information of text images. In this paper, we argue that scene text recognition is essentially a spatiotemporal prediction problem for its 2-D image inputs, and propose a convolution LSTM (ConvLSTM)-based scene text recognizer, namely, FACLSTM, i.e., Focused Attention ConvLSTM, where the spatial correlation of pixels is fully leveraged when performing sequential prediction with LSTM. Particularly, the attention mechanism is properly incorporated into an efficient ConvLSTM structure via the convolutional operations and additional character center masks are generated to help focus attention on right feature areas. The experimental results on benchmark datasets IIIT5K, SVT and CUTE demonstrate that our proposed FACLSTM performs competitively on the regular, low-resolution and noisy text images, and outperforms the state-of-the-art approaches on the curved text with large margins.

cs.CV↗

arXiv2018

Channel Attention and Multi-level Features Fusion for Single Image Super-Resolution

Convolutional neural networks (CNNs) have demonstrated superior performance in super-resolution (SR). However, most CNN-based SR methods neglect the different importance among feature channels or fail to take full advantage of the hierarchical features. To address these issues, this paper presents a novel recursive unit. Firstly, at the beginning of each unit, we adopt a compact channel attention mechanism to adaptively recalibrate the channel importance of input features. Then, the multi-level features, rather than only deep-level features, are extracted and fused. Additionally, we find that it will force our model to learn more details by using the learnable upsampling method (i.e., transposed convolution) only on residual branch (instead of using it both on residual branch and identity branch) while using the bicubic interpolation on the other branch. Analytic experiments show that our method achieves competitive results compared with the state-of-the-art methods and maintains faster speed as well.

cs.CV↗

arXiv2017

Nature of the growth of plasma electrolyte oxidation coating on Aluminum

The localized dielectric breakdown had been considered as the driving force for growth of plasma electrolytic oxide (PEO) coatings for several decades. However, the TEM study here reveals the dielectric breakdown behavior has little contribution for coating thickening. The presented evidences show the nature of PEO coating growth in all three consecutive stages I-III is ionic migration behavior inside the amorphous alumina layer (AAL) at the coating/matrix interface. The evolution of morphological characterizations in the PEO process is attributed to the interfacial reactions of AAL/alkaline (stage I), AAL/discharge-on-surface (stage II) and AAL/discharge-in-pore (stage III).

physics.app-ph↗

arXiv2017

Handwritten digit string recognition by combination of residual network and RNN-CTC

Recurrent neural network (RNN) and connectionist temporal classification (CTC) have showed successes in many sequence labeling tasks with the strong ability of dealing with the problems where the alignment between the inputs and the target labels is unknown. Residual network is a new structure of convolutional neural network and works well in various computer vision tasks. In this paper, we take advantage of the architectures mentioned above to create a new network for handwritten digit string recognition. First we design a residual network to extract features from input images, then we employ a RNN to model the contextual information within feature sequences and predict recognition results. At the top of this network, a standard CTC is applied to calculate the loss and yield the final results. These three parts compose an end-to-end trainable network. The proposed new architecture achieves the highest performances on ORAND-CAR-A and ORAND-CAR-B with recognition rates 89.75% and 91.14%, respectively. In addition, the experiments on a generated captcha dataset which has much longer string length show the potential of the proposed network to handle long strings.

cs.CV↗

arXiv2016

Control over emissivity of zero-static-power thermal emitters based on phase changing material GST

Controlling the emissivity of a thermal emitter has attracted growing interest with a view towards a new generation of thermal emission devices. So far, all demonstrations have involved sustained external electric or thermal consumption to maintain a desired emissivity. Here control over the emissivity of a thermal emitter consisting of a phase changing material Ge2Sb2Te5 (GST) film on top of a metal film is demonstrated. This thermal emitter shows broad wavelength-selective spectral emissivity in the mid-infrared. The peak emissivity approaches the ideal blackbody maximum and a maximum extinction ratio of above 10dB is attainable by switching GST between the crystalline and amorphous phases. By controlling the intermediate phases, the emissivity can be continuously tuned. This switchable, tunable, wavelength-selective and thermally stable thermal emitter will pave the way towards the ultimate control of thermal emissivity in the field of fundamental science as well as for energy-harvesting and thermal control applications, including thermophotovoltaics, light sources, infrared imaging and radiative coolers.

physics.optics↗