Chun Yang

Shi-Xue Zhang, Xiaobin Zhu, Jie-Bo Hou et al.

Segmentation-based methods have achieved great success for arbitrary shape text detection. However, separating neighboring text instances is still one of the most challenging problems due to the complexity of texts in scene images. In this paper, we propose an innovative Kernel Proposal Network (dubbed KPN) for arbitrary shape text detection. The proposed KPN can separate neighboring text instances by classifying different texts into instance-independent feature maps, meanwhile avoiding the complex aggregation process existing in segmentation-based arbitrary shape text detection methods. To be concrete, our KPN will predict a Gaussian center map for each text image, which will be used to extract a series of candidate kernel proposals (i.e., dynamic convolution kernel) from the embedding feature maps according to their corresponding keypoint positions. To enforce the independence between kernel proposals, we propose a novel orthogonal learning loss (OLL) via orthogonal constraints. Specifically, our kernel proposals contain important self-information learned by network and location information by position embedding. Finally, kernel proposals will individually convolve all embedding feature maps for generating individual embedded maps of text instances. In this way, our KPN can effectively separate neighboring text instances and improve the robustness against unclear boundaries. To our knowledge, our work is the first to introduce the dynamic convolution kernel strategy to efficiently and effectively tackle the adhesion problem of neighboring text instances in text detection. Experimental results on challenging datasets verify the impressive performance and efficiency of our method. The code and model are available at https://github.com/GXYM/KPN.

Shixiao Fan, Xuan Cheng, Xiaomin Wang et al. · gatech

Recently, researchers have shown an increased interest in the online knowledge distillation. Adopting an one-stage and end-to-end training fashion, online knowledge distillation uses aggregated intermediated predictions of multiple peer models for training. However, the absence of a powerful teacher model may result in the homogeneity problem between group peers, affecting the effectiveness of group distillation adversely. In this paper, we propose a novel online knowledge distillation method, \textbf{C}hannel \textbf{S}elf-\textbf{S}upervision for Online Knowledge Distillation (CSS), which structures diversity in terms of input, target, and network to alleviate the homogenization problem. Specifically, we construct a dual-network multi-branch structure and enhance inter-branch diversity through self-supervised learning, adopting the feature-level transformation and augmenting the corresponding labels. Meanwhile, the dual network structure has a larger space of independent parameters to resist the homogenization problem during distillation. Extensive quantitative experiments on CIFAR-100 illustrate that our method provides greater diversity than OKDDip and we also give pretty performance improvement, even over the state-of-the-art such as PCL. The results on three fine-grained datasets (StanfordDogs, StanfordCars, CUB-200-211) also show the significant generalization capability of our approach.

Chang Liu, Chun Yang, Hai-Bo Qin et al.

Scene text recognition is a popular topic and extensively used in the industry. Although many methods have achieved satisfactory performance for the close-set text recognition challenges, these methods lose feasibility in open-set scenarios, where collecting data or retraining models for novel characters could yield a high cost. For example, annotating samples for foreign languages can be expensive, whereas retraining the model each time when a novel character is discovered from historical documents costs both time and resources. In this paper, we introduce and formulate a new open-set text recognition task which demands the capability to spot and recognize novel characters without retraining. A label-to-prototype learning framework is also proposed as a baseline for the proposed task. Specifically, the framework introduces a generalizable label-to-prototype mapping function to build prototypes (class centers) for both seen and unseen classes. An open-set predictor is then utilized to recognize or reject samples according to the prototypes. The implementation of rejection capability over out-of-set characters allows automatic spotting of unknown characters in the incoming data stream. Extensive experiments show that our method achieves promising performance on a variety of zero-shot, close-set, and open-set text recognition datasets

Shi-Xue Zhang, Chun Yang, Xiaobin Zhu et al.

In arbitrary shape text detection, locating accurate text boundaries is challenging and non-trivial. Existing methods often suffer from indirect text boundary modeling or complex post-processing. In this paper, we systematically present a unified coarse-to-fine framework via boundary learning for arbitrary shape text detection, which can accurately and efficiently locate text boundaries without post-processing. In our method, we explicitly model the text boundary via an innovative iterative boundary transformer in a coarse-to-fine manner. In this way, our method can directly gain accurate text boundaries and abandon complex post-processing to improve efficiency. Specifically, our method mainly consists of a feature extraction backbone, a boundary proposal module, and an iteratively optimized boundary transformer module. The boundary proposal module consisting of multi-layer dilated convolutions will compute important prior information (including classification map, distance field, and direction field) for generating coarse boundary proposals while guiding the boundary transformer's optimization. The boundary transformer module adopts an encoder-decoder structure, in which the encoder is constructed by multi-layer transformer blocks with residual connection while the decoder is a simple multi-layer perceptron network (MLP). Under the guidance of prior information, the boundary transformer module will gradually refine the coarse boundary proposals via iterative boundary deformation. Furthermore, we propose a novel boundary energy loss (BEL) which introduces an energy minimization constraint and an energy monotonically decreasing constraint to further optimize and stabilize the learning of boundary refinement. Extensive experiments on publicly available and challenging datasets demonstrate the state-of-the-art performance and promising efficiency of our method.

Chang Liu, Chun Yang, Xu-Cheng Yin

The open-set text recognition task is an emerging challenge that requires an extra capability to cognize novel characters during evaluation. We argue that a major cause of the limited performance for current methods is the confounding effect of contextual information over the visual information of individual characters. Under open-set scenarios, the intractable bias in contextual information can be passed down to visual information, consequently impairing the classification performance. In this paper, a Character-Context Decoupling framework is proposed to alleviate this problem by separating contextual information and character-visual information. Contextual information can be decomposed into temporal information and linguistic information. Here, temporal information that models character order and word length is isolated with a detached temporal attention module. Linguistic information that models n-gram and other linguistic statistics is separated with a decoupled context anchor mechanism. A variety of quantitative and qualitative experiments show that our method achieves promising performance on open-set, zero-shot, and close-set text recognition datasets.

Zhan Shi, Xin Ding, Peng Ding et al.

Ship orientation angle prediction (SOAP) with optical remote sensing images is an important image processing task, which often relies on deep convolutional neural networks (CNNs) to make accurate predictions. This paper proposes a novel framework to reduce the model sizes and computational costs of SOAP models without harming prediction accuracy. First, a new SOAP model called Mobile-SOAP is designed based on MobileNetV2, achieving state-of-the-art prediction accuracy. Four tiny SOAP models are also created by replacing the convolutional blocks in Mobile-SOAP with four small-scale networks, respectively. Then, to transfer knowledge from Mobile-SOAP to four lightweight models, we propose a novel knowledge distillation (KD) framework termed SOAP-KD consisting of a novel feature-based guidance loss and an optimized synthetic samples-based knowledge transfer mechanism. Lastly, extensive experiments on the FGSC-23 dataset confirm the superiority of Mobile-SOAP over existing models and also demonstrate the effectiveness of SOAP-KD in improving the prediction performance of four specially designed tiny models. Notably, by using SOAP-KD, the test mean absolute error of the ShuffleNetV2x1.0-based model is only 8% higher than that of Mobile-SOAP, but its number of parameters and multiply-accumulate operations (MACs) are respectively 61.6% and 60.8% less.

Lei Chen, Haibo Qin, Shi-Xue Zhang et al.

In recent years, attention-based scene text recognition methods have been very popular and attracted the interest of many researchers. Attention-based methods can adaptively focus attention on a small area or even single point during decoding, in which the attention matrix is nearly one-hot distribution. Furthermore, the whole feature maps will be weighted and summed by all attention matrices during inference, causing huge redundant computations. In this paper, we propose an efficient attention-free Single-Point Decoding Network (dubbed SPDN) for scene text recognition, which can replace the traditional attention-based decoding network. Specifically, we propose Single-Point Sampling Module (SPSM) to efficiently sample one key point on the feature map for decoding one character. In this way, our method can not only precisely locate the key point of each character but also remove redundant computations. Based on SPSM, we design an efficient and novel single-point decoding network to replace the attention-based decoding network. Extensive experiments on publicly available benchmarks verify that our SPDN can greatly improve decoding efficiency without sacrificing performance.

Yan Xie, Changkui Mao, Changsong Wu et al.

As deep learning-based AI technologies gain momentum, the demand for general-purpose AI computing architectures continues to grow. While GPGPU-based architectures offer versatility for diverse AI workloads, they often fall short in efficiency and cost-effectiveness. Various Domain-Specific Architectures (DSAs) excel at particular AI tasks but struggle to extend across broader applications or adapt to the rapidly evolving AI landscape. M100 is Li Auto's response: a performant, cost-effective architecture for AI inference in Autonomous Driving (AD), Large Language Models (LLMs), and intelligent human interactions, domains crucial to today's most competitive automobile platforms. M100 employs a dataflow parallel architecture, where compiler-architecture co-design orchestrates not only computation but, more critically, data movement across time and space. Leveraging dataflow computing efficiency, our hardware-software co-design improves system performance while reducing hardware complexity and cost. M100 largely eliminates caching: tensor computations are driven by compiler- and runtime-managed data streams flowing between computing elements and on/off-chip memories, yielding greater efficiency and scalability than cache-based systems. Another key principle was selecting the right operational granularity for scheduling, issuing, and execution across compiler, firmware, and hardware. Recognizing commonalities in AI workloads, we chose the tensor as the fundamental data element. M100 demonstrates general AI computing capability across diverse inference applications, including UniAD (for AD) and LLaMA (for LLMs). Benchmarks show M100 outperforms GPGPU architectures in AD applications with higher utilization, representing a promising direction for future general AI computing.

Shi-Xue Zhang, Hongfa Wang, Xiaobin Zhu et al.

Video-language alignment is a crucial multi-modal task that benefits various downstream applications, e.g., video-text retrieval and video question answering. Existing methods either utilize multi-modal information in video-text pairs or apply global and local alignment techniques to promote alignment precision. However, these methods often fail to fully explore the spatio-temporal relationships among vision tokens within video and across different video-text pairs. In this paper, we propose a novel Spatio-Temporal Graph Transformer module to uniformly learn spatial and temporal contexts for video-language alignment pre-training (dubbed STGT). Specifically, our STGT combines spatio-temporal graph structure information with attention in transformer block, effectively utilizing the spatio-temporal contexts. In this way, we can model the relationships between vision tokens, promoting video-text alignment precision for benefiting downstream tasks. In addition, we propose a self-similarity alignment loss to explore the inherent self-similarity in the video and text. With the initial optimization achieved by contrastive learning, it can further promote the alignment accuracy between video and text. Experimental results on challenging downstream tasks, including video-text retrieval and video question answering, verify the superior performance of our method.

Chun Yang, Shicai Fan

The goal of a recommendation system is to model the relevance between each user and each item through the user-item interaction history, so that maximize the positive samples score and minimize negative samples. Currently, two popular loss functions are widely used to optimize recommender systems: the pointwise and the pairwise. Although these loss functions are widely used, however, there are two problems. (1) These traditional loss functions do not fit the goals of recommendation systems adequately and utilize prior knowledge information sufficiently. (2) The slow convergence speed of these traditional loss functions makes the practical application of various recommendation models difficult. To address these issues, we propose a novel loss function named Supervised Personalized Ranking (SPR) Based on Prior Knowledge. The proposed method improves the BPR loss by exploiting the prior knowledge on the interaction history of each user or item in the raw data. Unlike BPR, instead of constructing <user, positive item, negative item> triples, the proposed SPR constructs <user, similar user, positive item, negative item> quadruples. Although SPR is very simple, it is very effective. Extensive experiments show that our proposed SPR not only achieves better recommendation performance, but also significantly accelerates the convergence speed, resulting in a significant reduction in the required training time.

Junyi Zhang, Chang Liu, Chun Yang

In text recognition, complex glyphs and tail classes have always been factors affecting model performance. Specifically for Chinese text recognition, the lack of shape-awareness can lead to confusion among close complex characters. Since such characters are often tail classes that appear less frequently in the training-set, making it harder for the model to capture its shape information. Hence in this work, we propose a structure-aware network utilizing the hierarchical composition information to improve the recognition performance of complex characters. Implementation-wise, we first propose an auxiliary radical branch and integrate it into the base recognition network as a regularization term, which distills hierarchical composition information into the feature extractor. A Tree-Similarity-based weighting mechanism is then proposed to further utilize the depth information in the hierarchical representation. Experiments demonstrate that the proposed approach can significantly improve the performances of complex characters and tail characters, yielding a better overall performance. Code is available at https://github.com/Levi-ZJY/SAN.

Song-Lu Chen, Shu Tian, Jia-Wei Ma et al.

License plate detection is the first and essential step of the license plate recognition system and is still challenging in real applications, such as on-road scenarios. In particular, small-sized and oblique license plates, mainly caused by the distant and mobile camera, are difficult to detect. In this work, we propose a novel and applicable method for degraded license plate detection via vehicle-plate relation mining, which localizes the license plate in a coarse-to-fine scheme. First, we propose to estimate the local region around the license plate by using the relationships between the vehicle and the license plate, which can greatly reduce the search area and precisely detect very small-sized license plates. Second, we propose to predict the quadrilateral bounding box in the local region by regressing the four corners of the license plate to robustly detect oblique license plates. Moreover, the whole network can be trained in an end-to-end manner. Extensive experiments verify the effectiveness of our proposed method for small-sized and oblique license plates. Codes are available at https://github.com/chensonglu/LPD-end-to-end.

Zhiyu Fang, Shuai-Long Lei, Xiaobin Zhu et al.

Temporal Knowledge Graph (TKG) reasoning often involves completing missing factual elements along the timeline. Although existing methods can learn good embeddings for each factual element in quadruples by integrating temporal information, they often fail to infer the evolution of temporal facts. This is mainly because of (1) insufficiently exploring the internal structure and semantic relationships within individual quadruples and (2) inadequately learning a unified representation of the contextual and temporal correlations among different quadruples. To overcome these limitations, we propose a novel Transformer-based reasoning model (dubbed ECEformer) for TKG to learn the Evolutionary Chain of Events (ECE). Specifically, we unfold the neighborhood subgraph of an entity node in chronological order, forming an evolutionary chain of events as the input for our model. Subsequently, we utilize a Transformer encoder to learn the embeddings of intra-quadruples for ECE. We then craft a mixed-context reasoning module based on the multi-layer perceptron (MLP) to learn the unified representations of inter-quadruples for ECE while accomplishing temporal knowledge reasoning. In addition, to enhance the timeliness of the events, we devise an additional time prediction task to complete effective temporal information within the learned unified representation. Extensive experiments on six benchmark datasets verify the state-of-the-art performance and the effectiveness of our method.

Shi-Xue Zhang, Chun Yang, Xiaobin Zhu et al.

Scene text spotting is a challenging task, especially for inverse-like scene text, which has complex layouts, e.g., mirrored, symmetrical, or retro-flexed. In this paper, we propose a unified end-to-end trainable inverse-like antagonistic text spotting framework dubbed IATS, which can effectively spot inverse-like scene texts without sacrificing general ones. Specifically, we propose an innovative reading-order estimation module (REM) that extracts reading-order information from the initial text boundary generated by an initial boundary module (IBM). To optimize and train REM, we propose a joint reading-order estimation loss consisting of a classification loss, an orthogonality loss, and a distribution loss. With the help of IBM, we can divide the initial text boundary into two symmetric control points and iteratively refine the new text boundary using a lightweight boundary refinement module (BRM) for adapting to various shapes and scales. To alleviate the incompatibility between text detection and recognition, we propose a dynamic sampling module (DSM) with a thin-plate spline that can dynamically sample appropriate features for recognition in the detected text region. Without extra supervision, the DSM can proactively learn to sample appropriate features for text recognition through the gradient returned by the recognition module. Extensive experiments on both challenging scene text and inverse-like scene text datasets demonstrate that our method achieves superior performance both on irregular and inverse-like text spotting.

Zhiyu Fang, Jingyan Qin, Xiaobin Zhu et al.

Distinguished from traditional knowledge graphs (KGs), temporal knowledge graphs (TKGs) must explore and reason over temporally evolving facts adequately. However, existing TKG approaches still face two main challenges, i.e., the limited capability to model arbitrary timestamps continuously and the lack of rich inference patterns under temporal constraints. In this paper, we propose an innovative TKGE method (PTBox) via polynomial decomposition-based temporal representation and box embedding-based entity representation to tackle the above-mentioned problems. Specifically, we decompose time information by polynomials and then enhance the model's capability to represent arbitrary timestamps flexibly by incorporating the learnable temporal basis tensor. In addition, we model every entity as a hyperrectangle box and define each relation as a transformation on the head and tail entity boxes. The entity boxes can capture complex geometric structures and learn robust representations, improving the model's inductive capability for rich inference patterns. Theoretically, our PTBox can encode arbitrary time information or even unseen timestamps while capturing rich inference patterns and higher-arity relations of the knowledge base. Extensive experiments on real-world datasets demonstrate the effectiveness of our method.

Wanqi Wang, Chun Yang, Jianbo Shao et al.

Pediatric liver tumors are one of the most common solid tumors in pediatrics, with differentiation of benign or malignant status and pathological classification critical for clinical treatment. While pathological examination is the gold standard, the invasive biopsy has notable limitations: the highly vascular pediatric liver and fragile tumor tissue raise complication risks such as bleeding; additionally, young children with poor compliance require anesthesia for biopsy, increasing medical costs or psychological trauma. Although many efforts have been made to utilize AI in clinical settings, most researchers have overlooked its importance in pediatric liver tumors. To establish a non-invasive examination procedure, we developed a multi-stage deep learning (DL) framework for automated pediatric liver tumor diagnosis using multi-phase contrast-enhanced CT. Two retrospective and prospective cohorts were enrolled. We established a novel PKCP-MixUp data augmentation method to address data scarcity and class imbalance. We also trained a tumor detection model to extract ROIs, and then set a two-stage diagnosis pipeline with three backbones with ROI-masked images. Our tumor detection model has achieved high performance (mAP=0.871), and the first stage classification model between benign and malignant tumors reached an excellent performance (AUC=0.989). Final diagnosis models also exhibited robustness, including benign subtype classification (AUC=0.915) and malignant subtype classification (AUC=0.979). We also conducted multi-level comparative analyses, such as ablation studies on data and training pipelines, as well as Shapley-Value and CAM interpretability analyses. This framework fills the pediatric-specific DL diagnostic gap, provides actionable insights for CT phase selection and model design, and paves the way for precise, accessible pediatric liver tumor diagnosis.

Mengxue Yang, Chun Yang, Jiaqi Zhu et al.

Link prediction in knowledge graphs requires integrating structural information and semantic context to infer missing entities. While large language models offer strong generative reasoning capabilities, their limited exploitation of structural signals often results in structural sparsity and semantic ambiguity, especially under incomplete or zero-shot settings. To address these challenges, we propose SLiNT (Structure-aware Language model with Injection and coNtrastive Training), a modular framework that injects knowledge-graph-derived structural context into a frozen LLM backbone with lightweight LoRA-based adaptation for robust link prediction. Specifically, Structure-Guided Neighborhood Enhancement (SGNE) retrieves pseudo-neighbors to enrich sparse entities and mitigate missing context; Dynamic Hard Contrastive Learning (DHCL) introduces fine-grained supervision by interpolating hard positives and negatives to resolve entity-level ambiguity; and Gradient-Decoupled Dual Injection (GDDI) performs token-level structure-aware intervention while preserving the core LLM parameters. Experiments on WN18RR and FB15k-237 show that SLiNT achieves superior or competitive performance compared with both embedding-based and generation-based baselines, demonstrating the effectiveness of structure-aware representation learning for scalable knowledge graph completion.

Hao Liu, Chun Yang, Zhang xiaoxing et al.

Deep learning-based time series forecasting has found widespread applications. Recently, converting time series data into the frequency domain for forecasting has become popular for accurately exploring periodic patterns. However, existing methods often cannot effectively explore stationary information from complex intertwined frequency components. In this paper, we propose a simple yet effective Amplitude-Phase Reconstruct Network (APRNet) that models the inter-relationships of amplitude and phase, which prevents the amplitude and phase from being constrained by different physical quantities, thereby decoupling the distinct characteristics of signals for capturing stationary information. Specifically, we represent the multivariate time series input across sequence and channel dimensions, highlighting the correlation between amplitude and phase at multiple interaction frequencies. We propose a novel Kolmogorov-Arnold-Network-based Local Correlation (KLC) module to adaptively fit local functions using univariate functions, enabling more flexible characterization of stationary features across different amplitudes and phases. This significantly enhances the model's capability to capture time-varying patterns. Extensive experiments demonstrate the superiority of our APRNet against the state-of-the-arts (SOTAs).

Hao Liu, Chun Yang, Zhang xiaoxing et al.

Time series forecasting plays a significant role in finance, energy, meteorology, and IoT applications. Recent studies have leveraged the generalization capabilities of large language models (LLMs) to adapt to time series forecasting, achieving promising performance. However, existing studies focus on token-level modal alignment, instead of bridging the intrinsic modality gap between linguistic knowledge structures and time series data patterns, greatly limiting the semantic representation. To address this issue, we propose a novel Semantic-Enhanced LLM (SE-LLM) that explores the inherent periodicity and anomalous characteristics of time series to embed into the semantic space to enhance the token embedding. This process enhances the interpretability of tokens for LLMs, thereby activating the potential of LLMs for temporal sequence analysis. Moreover, existing Transformer-based LLMs excel at capturing long-range dependencies but are weak at modeling short-term anomalies in time-series data. Hence, we propose a plugin module embedded within self-attention that models long-term and short-term dependencies to effectively adapt LLMs to time-series analysis. Our approach freezes the LLM and reduces the sequence dimensionality of tokens, greatly reducing computational consumption. Experiments demonstrate the superiority performance of our SE-LLM against the state-of-the-art (SOTA) methods.

Chun Yang

In this work, we aim to consider the application of contrastive learning in the scenario of the recommendation system adequately, making it more suitable for recommendation task. We propose a learning paradigm called supervised contrastive learning(SCL) to support the graph convolutional neural network. Specifically, we will calculate the similarity between different nodes in user side and item side respectively during data preprocessing, and then when applying contrastive learning, not only will the augmented views be regarded as the positive samples, but also a certain number of similar samples will be regarded as the positive samples, which is different with SimCLR that treats other samples in a batch as negative samples. We apply SCL on the most advanced LightGCN. In addition, in order to consider the uncertainty of node interaction, we also propose a new data augment method called node replication. Empirical research and ablation study on Gowalla, Yelp2018, Amazon-Book datasets prove the effectiveness of SCL and node replication, which improve the accuracy of recommendations and robustness to interactive noise.

Zhiyu Fang, Xiaobin Zhu, Chun Yang et al.

Learning a common latent embedding by aligning the latent spaces of cross-modal autoencoders is an effective strategy for Generalized Zero-Shot Classification (GZSC). However, due to the lack of fine-grained instance-wise annotations, it still easily suffer from the domain shift problem for the discrepancy between the visual representation of diversified images and the semantic representation of fixed attributes. In this paper, we propose an innovative autoencoder network by learning Aligned Cross-Modal Representations (dubbed ACMR) for GZSC. Specifically, we propose a novel Vision-Semantic Alignment (VSA) method to strengthen the alignment of cross-modal latent features on the latent subspaces guided by a learned classifier. In addition, we propose a novel Information Enhancement Module (IEM) to reduce the possibility of latent variables collapse meanwhile encouraging the discriminative ability of latent variables. Extensive experiments on publicly available datasets demonstrate the state-of-the-art performance of our method.

Chun Yang

Considering the models that apply the contextual information of time-series data could improve the fault diagnosis performance, some neural network structures such as RNN, LSTM, and GRU were proposed to model the fault diagnosis effectively. However, these models are restricted by their serial computation and hence cannot achieve high diagnostic efficiency. Also the parallel CNN is difficult to implement fault diagnosis in an efficient way because it requires larger convolution kernels or deep structure to achieve long-term feature extraction capabilities. Besides, BERT model applies absolute position embedding to introduce contextual information to the model, which would bring noise to the raw data and therefore cannot be applied to fault diagnosis directly. In order to address the above problems, a fault diagnosis model named deep parallel time-series relation network(DPTRN) has been proposed in this paper. There are mainly three advantages for DPTRN: (1) Our proposed time relationship unit is based on full multilayer perceptron(MLP) structure, therefore, DPTRN performs fault diagnosis in a parallel way and improves computing efficiency significantly. (2) By improving the absolute position embedding, our novel decoupling position embedding unit could be applied on the fault diagnosis directly and learn contextual information. (3) Our proposed DPTRN has obvious advantage in feature interpretability. We confirm the effect of the proposed method on four datasets, and the results show the effectiveness, efficiency and interpretability of the proposed DPTRN model.

Shi-Xue Zhang, Xiaobin Zhu, Chun Yang et al.

Arbitrary shape text detection is a challenging task due to the high complexity and variety of scene texts. In this work, we propose a novel adaptive boundary proposal network for arbitrary shape text detection, which can learn to directly produce accurate boundary for arbitrary shape text without any post-processing. Our method mainly consists of a boundary proposal model and an innovative adaptive boundary deformation model. The boundary proposal model constructed by multi-layer dilated convolutions is adopted to produce prior information (including classification map, distance field, and direction field) and coarse boundary proposals. The adaptive boundary deformation model is an encoder-decoder network, in which the encoder mainly consists of a Graph Convolutional Network (GCN) and a Recurrent Neural Network (RNN). It aims to perform boundary deformation in an iterative way for obtaining text instance shape guided by prior information from the boundary proposal model. In this way, our method can directly and efficiently generate accurate text boundaries without complex post-processing. Extensive experiments on publicly available datasets demonstrate the state-of-the-art performance of our method.

Yujie Cui, Chun Yang, Xu Cheng

Caches have been used to construct various types of covert and side channels to leak information. Most existing cache channels exploit the timing difference between cache hits and cache misses. However, we introduce a new and broader classification of cache covert channel attacks: Hit+Miss, Hit+Hit, and Miss+Miss. We highlight that cache misses for cache lines in different states may have more significant time differences, and these can be used as timing channels. Based on this classification, we propose a new stable and stealthy Miss+Miss cache channel. Write-back caches are widely deployed in modern processors. This paper presents in detail a way in which replacement latency differences can be used to construct timing-based channels (called WB channels) to leak information in a write-back cache. Any modification to a cache line by a sender will set it to the dirty state, and the receiver can observe this through measuring the latency of replacing this cache set. We also demonstrate how senders could exploit a different number of dirty cache lines in a cache set to improve transmission bandwidth with symbols encoding multiple bits. The peak transmission bandwidths of the WB channels in commercial systems can vary between 1300 and 4400~kbps per cache set in a hyper-threaded setting without shared memory between the sender and the receiver. In contrast to most existing cache channels, which always target specific memory addresses, the new WB channels focus on the cache set and cache line states, making it difficult for the channel to be disturbed by other processes on the core, and they can still work in a cache using a random replacement policy. We also analyzed the stealthiness of WB channels from the perspective of the number of cache loads and cache miss rates. We discuss and evaluate possible defenses. The paper finishes by discussing various forms of side-channel attack.

Chun Yang, Franz Rottensteiner, Christian Heipke

Land use as contained in geospatial databases constitutes an essential input for different applica-tions such as urban management, regional planning and environmental monitoring. In this paper, a hierarchical deep learning framework is proposed to verify the land use information. For this purpose, a two-step strategy is applied. First, given high-resolution aerial images, the land cover information is determined. To achieve this, an encoder-decoder based convolutional neural net-work (CNN) is proposed. Second, the pixel-wise land cover information along with the aerial images serves as input for another CNN to classify land use. Because the object catalogue of geospatial databases is frequently constructed in a hierarchical manner, we propose a new CNN-based method aiming to predict land use in multiple levels hierarchically and simultaneously. A so called Joint Optimization (JO) is proposed where predictions are made by selecting the hier-archical tuple over all levels which has the maximum joint class scores, providing consistent results across the different levels. The conducted experiments show that the CNN relying on JO outperforms previous results, achieving an overall accuracy up to 92.5%. In addition to the individual experiments on two test sites, we investigate whether data showing different characteristics can improve the results of land cover and land use classification, when processed together. To do so, we combine the two datasets and undertake some additional experiments. The results show that adding more data helps both land cover and land use classification, especially the identification of underrepre-sented categories, despite their different characteristics.

Zan-Xia Jin, Heran Wu, Chun Yang et al.

Text-based visual question answering (VQA) requires to read and understand text in an image to correctly answer a given question. However, most current methods simply add optical character recognition (OCR) tokens extracted from the image into the VQA model without considering contextual information of OCR tokens and mining the relationships between OCR tokens and scene objects. In this paper, we propose a novel text-centered method called RUArt (Reading, Understanding and Answering the Related Text) for text-based VQA. Taking an image and a question as input, RUArt first reads the image and obtains text and scene objects. Then, it understands the question, OCRed text and objects in the context of the scene, and further mines the relationships among them. Finally, it answers the related text for the given question through text semantic matching and reasoning. We evaluate our RUArt on two text-based VQA benchmarks (ST-VQA and TextVQA) and conduct extensive ablation studies for exploring the reasons behind RUArt's effectiveness. Experimental results demonstrate that our method can effectively explore the contextual information of the text and mine the stable relationships between the text and objects.

Po-Heng Chen, Zhao-Xu Luo, Zu-Kuan Huang et al.

Feature descriptor matching is a critical step is many computer vision applications such as image stitching, image retrieval and visual localization. However, it is often affected by many practical factors which will degrade its performance. Among these factors, illumination variations are the most influential one, and especially no previous descriptor learning works focus on dealing with this problem. In this paper, we propose IF-Net, aimed to generate a robust and generic descriptor under crucial illumination changes conditions. We find out not only the kind of training data important but also the order it is presented. To this end, we investigate several dataset scheduling methods and propose a separation training scheme to improve the matching accuracy. Further, we propose a ROI loss and hard-positive mining strategy along with the training scheme, which can strengthen the ability of generated descriptor dealing with large illumination change conditions. We evaluate our approach on public patch matching benchmark and achieve the best results compared with several state-of-the-arts methods. To show the practicality, we further evaluate IF-Net on the task of visual localization under large illumination changes scenes, and achieves the best localization accuracy.

Shi-Xue Zhang, Xiaobin Zhu, Jie-Bo Hou et al.

Arbitrary shape text detection is a challenging task due to the high variety and complexity of scenes texts. In this paper, we propose a novel unified relational reasoning graph network for arbitrary shape text detection. In our method, an innovative local graph bridges a text proposal model via Convolutional Neural Network (CNN) and a deep relational reasoning network via Graph Convolutional Network (GCN), making our network end-to-end trainable. To be concrete, every text instance will be divided into a series of small rectangular components, and the geometry attributes (e.g., height, width, and orientation) of the small components will be estimated by our text proposal model. Given the geometry attributes, the local graph construction model can roughly establish linkages between different text components. For further reasoning and deducing the likelihood of linkages between the component and its neighbors, we adopt a graph-based network to perform deep relational reasoning on local graphs. Experiments on public available datasets demonstrate the state-of-the-art performance of our method.

Daliang Xu, Dongwei Chen, Chun Yang et al.

Use-After-Free vulnerabilities, allowing the attacker to access unintended memory via dangling pointers, are more threatening. However, most detection schemes can only detect dangling pointers and invalid them, but not provide a tolerance mechanism to repair the errors at runtime. Also, these techniques obtain and manage the metadata inefficiently with complex structures and too much scan (sweep). The goal of this paper is to use compiler instrumentation to eliminate dangling pointers automatically and efficiently. In this paper, we observe that most techniques lack accurate efficient pointer graph metadata maintaining methods, so they need to scan the log to reduce the redundancy and sweep the whole address space to find dangling pointers. Also, they lack a direct, efficiently obtaining metadata approach. The key insight of this paper is that a unique identifier can be used as a key to a hash or direct-map algorithm. Thus, this paper maintains the same implicit identifier with each memory object and its corresponding referent. Associating the unique ID with metadata for memory objects, obtaining and managing the pointer graph metadata can be efficiently. Therefore, with the delayed free technique adopted into C/C++, we present the DangKiller as a novel and lightweight dangling pointer elimination solution. We first demonstrate the MinFat Pointer, which can calculate unique implicit ID for each object and pointer quickly, and use hash algorithm to obtain metadata. Secondly, we propose the Log Cache and Log Compression mechanism based on the ID to decrease the redundancy of dangling pointer candidates. Coupled with the Address Tagging architecture on an ARM64 system, our experiments show that the DangKiller can eliminate use-after-free vulnerabilities at only 11% and 3% runtime overheads for the SPEC CPU2006 and 2017 benchmarks respectively, except for unique cases.

Dongwei Chen, Daliang Xu, Dong Tong et al.

Memory spatial errors, i.e., buffer overflow vulnerabilities, have been a well-known issue in computer security for a long time and remain one of the root causes of exploitable vulnerabilities. Most of the existing mitigation tools adopt a fail-stop strategy to protect programs from intrusions, which means the victim program will be terminated upon detecting a memory safety violation. Unfortunately, the fail-stop strategy harms the availability of software. In this paper, we propose Saturation Memory Access (SMA), a memory spatial error mitigation mechanism that prevents out-of-bounds access without terminating a program. SMA is based on a key observation that developers generally do not rely on out-of-bounds accesses to implement program logic. SMA modifies dynamic memory allocators and adds paddings to objects to form an enlarged object boundary. By dynamically correcting all the out-of-bounds accesses to operate on the enlarged protecting boundaries, SMA can tolerate out-of-bounds accesses. For the sake of compatibility, we chose tagged pointers to record the boundary metadata of a memory object in the pointer itself, and correct the address upon detecting out-of-bounds access. We have implemented the prototype of SMA on LLVM 10.0. Our results show that our compiler enables the programs to execute successfully through buffer overflow attacks. Experiments on MiBench show that our prototype incurs an overhead of 78\%. Further optimizations would require ISA supports.

Xinjie Li, Chun Yang, Songlu Chen et al.

Naturally, fine-grained recognition, e.g., vehicle identification or bird classification, has specific hierarchical labels, where fine categories are always harder to be classified than coarse categories. However, most of the recent deep learning based methods neglect the semantic structure of fine-grained objects and do not take advantage of the traditional fine-grained recognition techniques (e.g. coarse-to-fine classification). In this paper, we propose a novel framework with a two-branch network (coarse branch and fine branch), i.e., semantic bilinear pooling, for fine-grained recognition with a hierarchical label tree. This framework can adaptively learn the semantic information from the hierarchical levels. Specifically, we design a generalized cross-entropy loss for the training of the proposed framework to fully exploit the semantic priors via considering the relevance between adjacent levels and enlarge the distance between samples of different coarse classes. Furthermore, our method leverages only the fine branch when testing so that it adds no overhead to the testing time. Experimental results show that our proposed method achieves state-of-the-art performance on four public datasets.

Michael Ying Yang, Wentong Liao, Chun Yang et al.

With rapidly increasing deployment of surveillance cameras, the reliable methods for automatically analyzing the surveillance video and recognizing special events are demanded by different practical applications. This paper proposes a novel effective framework for security event analysis in surveillance videos. First, convolutional neural network (CNN) framework is used to detect objects of interest in the given videos. Second, the owners of the objects are recognized and monitored in real-time as well. If anyone moves any object, this person will be verified whether he/she is its owner. If not, this event will be further analyzed and distinguished between two different scenes: moving the object away or stealing it. To validate the proposed approach, a new video dataset consisting of various scenarios is constructed for more complex tasks. For comparison purpose, the experiments are also carried out on the benchmark databases related to the task on abandoned luggage detection. The experimental results show that the proposed approach outperforms the state-of-the-art methods and effective in recognizing complex security events.

Chun Yang, Xu-Cheng Yin, Zejun Li et al.

Recognizing text in the wild is a really challenging task because of complex backgrounds, various illuminations and diverse distortions, even with deep neural networks (convolutional neural networks and recurrent neural networks). In the end-to-end training procedure for scene text recognition, the outputs of deep neural networks at different iterations are always demonstrated with diversity and complementarity for the target object (text). Here, a simple but effective deep learning method, an adaptive ensemble of deep neural networks (AdaDNNs), is proposed to simply select and adaptively combine classifier components at different iterations from the whole learning system. Furthermore, the ensemble is formulated as a Bayesian framework for classifier weighting and combination. A variety of experiments on several typical acknowledged benchmarks, i.e., ICDAR Robust Reading Competition (Challenge 1, 2 and 4) datasets, verify the surprised improvement from the baseline DNNs, and the effectiveness of AdaDNNs compared with the recent state-of-the-art methods.

Xu-Cheng Yin, Chun Yang, Hong-Wei Hao

Classifier ensemble generally should combine diverse component classifiers. However, it is difficult to give a definitive connection between diversity measure and ensemble accuracy. Given a list of available component classifiers, how to adaptively and diversely ensemble classifiers becomes a big challenge in the literature. In this paper, we argue that diversity, not direct diversity on samples but adaptive diversity with data, is highly correlated to ensemble accuracy, and we propose a novel technology for classifier ensemble, learning to diversify, which learns to adaptively combine classifiers by considering both accuracy and diversity. Specifically, our approach, Learning TO Diversify via Weighted Kernels (L2DWK), performs classifier combination by optimizing a direct but simple criterion: maximizing ensemble accuracy and adaptive diversity simultaneously by minimizing a convex loss function. Given a measure formulation, the diversity is calculated with weighted kernels (i.e., the diversity is measured on the component classifiers' outputs which are kernelled and weighted), and the kernel weights are automatically learned. We minimize this loss function by estimating the kernel weights in conjunction with the classifier weights, and propose a self-training algorithm for conducting this convex optimization procedure iteratively. Extensive experiments on a variety of 32 UCI classification benchmark datasets show that the proposed approach consistently outperforms state-of-the-art ensembles such as Bagging, AdaBoost, Random Forests, Gasen, Regularized Selective Ensemble, and Ensemble Pruning via Semi-Definite Programming.