Video description aims to generate descriptive natural language for videos.Inspired from the deep neural network(DNN) used in the machine translation,the video description(VD) task applies the convolutional neural network(CNN) to extracting video features and the long short-term memory(LSTM) to generating descriptions.However,some models generate incorrect words and syntax.The reason may because that the previous models only apply LSTM to generate sentences,which learn insufficient linguistic information.In order to solve this problem,an end-to-end DNN model incorporated subject,verb and object(SVO) supervision is proposed.Experimental results on a publicly available dataset,i.e.Youtube2 Text,indicate that our model gets a 58.4% consensus-based image description evaluation(CIDEr) value.It outperforms the mean pool and video description with first feed(VD-FF) models,demonstrating the effectiveness of SVO supervision.
Utilizing a three-particle W state, we come up with a protocol for the teleportation of an unknown two-particle entangled state. It is shown that the teleportation can be deterministically and exactly realized. Moreover, two-particle entanglement teleportation is generalized to a system consisting of many particles via a three-particle W state and a multi-particle W state, respectively. All unitary transformations performed by the receiver are given in a concise formula.
Yushu Tibetan Autonomous Prefecture, an area located in the Qinghai-Tibet Plateau, is an area very sensitive to global climate change. Due to impacts from climate change and human disturbances, grassland vegetation in the area has been degraded and desertification has been expanding. Ecosystems in the area are very sensitive and fragile and ecological problems have become increasingly serious in the area, resulting in an adverse effect on the local socio-economic development and environment of Qinghai province. Using data gathered from Landsat TM/ETM images for 1987, 1997 and 2007, we analyzed landscape patterns across Yushu Prefecture. Spatial structure indices indicated that： （i） the area of grassland has significantly decreased in the form of degradation and conversion from grassland into bare land and farmland; （ii） grassy vegetation patches changed into fragmented and isolated patches; （iii） the main landscapes in Yushu Prefecture are grasslands, forests and rivers; （iv） patches of grass have reiatively high connectivity; and （v） landscape change is significantly correlated with human activities and climate change. This study provides a strong theoretical and technical basis for policy-making regarding environmental protection of and management in Yushu Prefecture of Qinghai Province.
