The dominant video question answering methods are based on fine-grained representation or model-specific attention mechanism. They usually process video and question separately, then feed the representations of different modalities into following late fusion networks. Although these methods use information of one modality to boost the other, they neglect to integrate correlations of both inter- and intra-modality in an uniform module. We propose a deep heterogeneous graph alignment network over the video shots and question words. Furthermore, we explore the network architecture from four steps: representation, fusion, alignment, and reasoning. Within our network, the inter- and intra-modality information can be aligned and interacted simultaneously over the heterogeneous graph and used for cross-modal reasoning. We evaluate our method on three benchmark datasets and conduct extensive ablation study to the effectiveness of the network architecture. Experiments show the network to be superior in quality.

Semi-supervised domain adaptation (SSDA) is a novel branch of machine learning that scarce labeled target examples are available, compared with unsupervised domain adaptation. To make effective use of these additional data so as to bridge the domain gap, one possible way is to generate adversarial examples, which are images with additional perturbations, between the two domains and fill the domain gap. Adversarial training has been proven to be a powerful method for this purpose. However, the traditional adversarial training adds noises in arbitrary directions, which is inefficient to migrate between domains, or generate directional noises from the source to target domain and reverse. In this work, we devise a general bidirectional adversarial training method and employ gradient to guide adversarial examples across the domain gap, i.e., the Adaptive Adversarial Training (AAT) for source to target domain and Entropy-penalized Virtual Adversarial Training (E-VAT) for target to source domain. Particularly, we devise a Bidirectional Adversarial Training (BiAT) network to perform diverse adversarial trainings jointly. We evaluate the effectiveness of BiAT on three benchmark datasets and experimental results demonstrate the proposed method achieves the state-of-the-art.

This paper focuses on the query-focused video summarization, which is an extended task of video summarization and aims to automatically generate user-oriented summary by highlighting frames/shots relevant to the query. This task is different from traditional video summarization in paying attention to users' subjectivity through queries. Diversity is a recognized important property in video summarization. However, existing methods only consider diversity as the dissimilarity between frames/shots which is far from user-oriented summarization. Users' different understandings of video should be an important source of diversity, reflected in the process of eliminating query-unrelated redundancy. To this end, this paper explores user-diversified & query-focused video summarization via a well-devised hierarchical variational network called HVN. HVN has three distinctive characteristics: (i) it has a hierarchical structure to model query-related long-range temporal dependency; (ii) it employs diverse attention mechanisms to encode query-related and context-important information and makes them balanced; (iii) it employs a multilevel self-attention module and a variational autoencoder module to add user-oriented diversity and stochastic factors. Experimental results demonstrate that HVN not only outperforms the state-of-the-arts but also improves the user-oriented diversity to some extent.

With mobile WiMAX being widely deployed, providing a feasible solution for location service in mobile WiMAX is imperative and significant. In this paper, we propose a novel mobile WiMAX positioning scheme, which includes four steps: downlink preamble fast detection (DPFD), positioning algorithm selection mechanism based on the number of detected preambles, media access control (MAC) message signaling, and location calculation. In the first step, neighbor information carried by MOBNBR-ADV is introduced to decrease the neighbor BSs' preamble detection time. Meanwhile, Uplink Ranging Assistant Downlink-Time Difference of Arrival (URAD-TDoA)/uplink ranging assistant-cell ID (URA-CelllD) are proposed in selection mechanism for two/one preambles detected. Furthermore, new messages and detail message signaling for positioning are designed in MAC layer. By simulations, we show that the proposed positioning scheme can be a promising candidate for location service in mobile WiMAX systems.

This paper focuses on routing tree construction problem and its influence on the performance of utilizing centralized scheduling in IEEE 802.16 mesh networks. We apply three routing tree construction algorithms, namely, hop minimization modulation maximization (HMMM), energy/bit minimization (EbM), and interference minimization (IM) routing tree constructions with nodes spaced randomly. Furthermore, a novel multi-channel centralized scheduling algorithm with spatial reuse is proposed and its performance with the three routing tree construction algorithms is evaluated under variance of link capacity induced by co-channel interference. Simulation results show that the routing tree constructed by the EbM algorithm outperforms the two others. Therefore, EbM routing tree construction algorithm with multichannel scheduling algorithm with spatial reuse is highly efficient for IEEE 802.16 mesh networks.

Mobile WiMAX is one of the most promising technologies for broadband wireless communication. The IEEE 802.16e standard for Mobile WiMAX, the enhanced version of the IEEE 802.16 standard with mobility support, defines the implementation of hard handover is mandatory. However, the long interruption of hard handover is horrible for delay and packet loss sensitive real-time applications such as IPTV, VoIP, SatTV. To solve this problem without many changes to 802.16e equipments, we propose an enhanced link-layer QoS aware handover scheme - passport handover to accelerate handover process. We also propose a connection CID assignment strategy to avoid confliction of CIDs of handing over services with that of ongoing services in the target BS. Simulation results prove that the scheme makes real-time downlink data transmission interruption so brief that users will not perceive the interruption. And it also greatly reduces uplink data transmission delay and packet loss probability.