Cross-view action recognition is a challenging research field for human motion analysis. Appearance-based features are not credible if the viewpoint changes. In this paper, a new framework is proposed for cross-view action recognition by topic based knowledge transfer. First, Spatio-temporal descriptors are extracted from the action videos and each video is modeled by a bag of visual words (BoVW) based on the codebook constructed by the k-means cluster algorithm. Second, Latent Dirichlet Allocation (LDA) is employed to assign topics for the BoVW representation. The topic distribution of visual words (ToVW) is normalized and taken to be the feature vector. Third, in order to bridge different views, we transform ToVW into bilingual ToVW by constructing bilingual dictionaries, which guarantee that the same action has the same representation from different views. We demonstrate the effectiveness of the proposed algorithm on the IXMAS multi-view dataset.

In many radio frequency identification (RFID) applications, the reader identifies the tags in its scope repeatedly. For these applications, many algorithms, such as an adaptive binary splitting algorithm (ABS), a single resolution blocking ABS (SRB), a pair resolution blocking ABS (PRB) and a dynamic blocking ABS (DBA) have been proposed. All these algorithms require the staying tags to reply with their IDs to be recognized by the reader. However, the IDs of the staying tags are stored in the reader in the last identification round. The reader can verify the existence of these tags when identifying them. Thus, we propose an anti-collision algorithm with short reply for RFID tag identification (ACSR). In ACSR, each staying tag emits a short reply to indicate its continued existence. Therefore, the data amount transmitted by staying tags is reduced significantly. The identification rate of ACSR is analyzed in this paper. Finally, simulation and analysis results show that ACSR greatly outperforms ABS, SRB and DBA in terms of the identification rate and average amount of data transmitted by a tag.

A variety of ZnO belt-like structures were synthesized by the heat treatment of ZnS substrates with Ga droplets in the air, and their morphological and structural properties were investigated. Three types of ZnO belts with flat surfaces of (20), (100) and (23) were obtained. As comparison, the ZnO crystal growth was examined by the thermal oxidation of ZnS only. These results highlight the promise of the heat treatment with Ga in the synthesis of oxide nanostructures.

In this paper, the outage probability and diversity order of opportunistic decode-and-forward (DF) cooperation are analyzed under Rayleigh fading channels, where the impacts of channel estimation error, relay selection feedback delay and the availability of the direct link between the source and the destination are considered comprehensively. The closed-form expressions of outage probability in the high signal-to-noise ratio (SNR) region are derived as well as the diversity order. The theoretical results demonstrate that the achievable diversity order is zero when channel estimation error exists, and this conclusion holds no matter whether the direct link is available, even if the relay selection feedback is delay-free. For the perfect channel estimation scenario, the achievable diversity order is related to the potential relay number K, the channel delay correlation coefficient ρd and the availability of the direct link. If relay selection feedback is not delayed, i.e., ρd = 1, the diversity order is K when the direct link is blocked, and it becomes K+1 when the direct link is available. For delayed relay selection feedback, i.e., ρd < 1, the diversity order achievable is only related to the availability of the direct link. In this case, if the direct link does not exist, the diversity order is 1, otherwise the diversity order of 2 can be obtained. Simulation results verify the analytical results of outage probability and diversity order.