Xing ZHU Yuxuan LIU Lingyu LIANG Tao WANG Zuoyong LI Qiaoming DENG Yubo LIU
Recently, many AI-aided layout design systems are developed to reduce tedious manual intervention based on deep learning. However, most methods focus on a specific generation task. This paper explores a challenging problem to obtain multiple layout design generation (LDG), which generates floor plan or urban plan from a boundary input under a unified framework. One of the main challenges of multiple LDG is to obtain reasonable topological structures of layout generation with irregular boundaries and layout elements for different types of design. This paper formulates the multiple LDG task as an image-to-image translation problem, and proposes a conditional generative adversarial network (GAN), called LDGAN, with adaptive modules. The framework of LDGAN is based on a generator-discriminator architecture, where the generator is integrated with conditional convolution constrained by the boundary input and the attention module with channel and spatial features. Qualitative and quantitative experiments were conducted on the SCUT-AutoALP and RPLAN datasets, and the comparison with the state-of-the-art methods illustrate the effectiveness and superiority of the proposed LDGAN.
Wenting CHANG Jintao WANG Bo AI
A scheme that modulates the training sequence is proposed to support two-layer data streams in the time domain synchronous orthogonal frequency division multiplex (TDS-OFDM) systems. A theoretical analysis and computer simulation show that the proposed scheme works well and that the two layer data streams are compatible with each other.
Tao WANG Hongchang CHEN Chao QI
Software-defined networking (SDN) has rapidly emerged as a promising new technology for future networks and gained considerable attention from both academia and industry. However, due to the separation between the control plane and the data plane, the SDN controller can easily become the target of denial-of service (DoS) attacks. To mitigate DoS attacks in OpenFlow networks, our solution, MinDoS, contains two key techniques/modules: the simplified DoS detection module and the priority manager. The proposed architecture sends requests into multiple buffer queues with different priorities and then schedules the processing of these flow requests to ensure better controller protection. The results show that MinDoS is effective and adds only minor overhead to the entire SDN/OpenFlow infrastructure.
Dengbao DU Jintao WANG Jun WANG Ke GONG Zhixing YANG
A differential inter-symbol interference (ISI) cancellation method for time domain synchronous orthogonal frequency division multiplexing (TDS-OFDM) systems is proposed. The differential output of an OFDM system can greatly reduce the impact of ISI in the frequency domain and it constructs a convolutional structure, thus the Viterbi decoding algorithm can be used to recover the transmitted information from the output signal. Simulation results show the effectiveness of the proposed method.
Wocheng XIAO Lingyu LIANG Jianyong CHEN Tao WANG
Video text detection (VTD) aims to localize text instances in videos, which has wide applications for downstream tasks. To deal with the variances of different scenes and text instances, multiple models and feature fusion strategies were typically integrated in existing VTD methods. A VTD method consisting of sophisticated components can efficiently improve detection accuracy, but may suffer from a limitation for real-time applications. This paper aims to achieve real-time VTD with an adaptive lightweight end-to-end framework. Different from previous methods that represent text in a spatial domain, we model text instances in the Fourier domain. Specifically, we propose a scale-aware Fourier Contour Embedding method, which not only models arbitrary shaped text contours of videos as compact signatures, but also adaptively select proper scales for features in a backbone in the training stage. Then, we construct VTD-FCENet to achieve real-time VTD, which encodes temporal correlations of adjacent frames with scale-aware FCE in a lightweight and adaptive manner. Quantitative evaluations were conducted on ICDAR2013 Video, Minetto and YVT benchmark datasets, and the results show that our VTD-FCENet not only obtains the state-of-the-arts or competitive detection accuracy, but also allows real-time text detection on HD videos simultaneously.