A lightweight multi-layer residual temporal convolutional network model(RTCN)is proposed to target the highly complex kinematics and temporal correlation of human motion.RTCN uses 1-D convolution to efficiently obtain...A lightweight multi-layer residual temporal convolutional network model(RTCN)is proposed to target the highly complex kinematics and temporal correlation of human motion.RTCN uses 1-D convolution to efficiently obtain the spatial structure information of human motion and extract the correlation in the time series of human motion.The residual structure is applied to the proposed network model to alleviate the problem of gradient disappearance in the deep network.Experiments on the Human 3.6M dataset demonstrate that the proposed method effectively reduces the errors of motion prediction compared with previous methods,especially of long-term prediction.展开更多
Using 1200 CPUs of the National Supercomputer TH-A1 and a parallel integral algorithm based on the 3500th-order Taylor expansion and the 4180-digit multiple precision data,we have done a reliable simulation of chaotic...Using 1200 CPUs of the National Supercomputer TH-A1 and a parallel integral algorithm based on the 3500th-order Taylor expansion and the 4180-digit multiple precision data,we have done a reliable simulation of chaotic solution of Lorenz equation in a rather long interval 0 t 10000 LTU(Lorenz time unit).Such a kind of mathematically reliable chaotic simulation has never been reported.It provides us a numerical benchmark for mathematically reliable long-term prediction of chaos.Besides,it also proposes a safe method for mathematically reliable simulations of chaos in a finite but long enough interval.In addition,our very fine simulations suggest that such a kind of mathematically reliable long-term prediction of chaotic solution might have no physical meanings,because the inherent physical micro-level uncertainty due to thermal fluctuation might quickly transfer into macroscopic uncertainty so that trajectories for a long enough time would be essentially uncertain in physics.展开更多
The latest video coding standard High Efficiency Video Coding （HEVC） can achieve much higher coding efficiency than previous video coding standards. Particularly, by exploiting the hierarchical B-picture prediction ...The latest video coding standard High Efficiency Video Coding （HEVC） can achieve much higher coding efficiency than previous video coding standards. Particularly, by exploiting the hierarchical B-picture prediction structure, temporal redundancy among neighbor frarnes is eliminated remarkably well. In practice, videos available to consumers usually contain many repeated shots, such as TV series, movies, and talk shows. According to our observations, when these videos are encoded by HEVC with the hierarchical B-picture structure, the temporal correlation in each shot is well exploited. However, the long-term correlation between repeated shots has not been used. We propose a long-term prediction （LTP） scheme to use the long-term temporal correlation between correlated shots in a video. The long-term reference （LTR） frames of a source video are chosen by clustering similar shots and extracting the representative frames, and a modified hierarchical B-picture coding structure based on an LTR frame is introduced to support long-term temporal prediction. An adaptive quantization method is further designed for LTR frames to improve the overall video coding efficiency. Experimental results show that up to 22.86% coding gain can be achieved using the new coding scheme.展开更多
文摘A lightweight multi-layer residual temporal convolutional network model(RTCN)is proposed to target the highly complex kinematics and temporal correlation of human motion.RTCN uses 1-D convolution to efficiently obtain the spatial structure information of human motion and extract the correlation in the time series of human motion.The residual structure is applied to the proposed network model to alleviate the problem of gradient disappearance in the deep network.Experiments on the Human 3.6M dataset demonstrate that the proposed method effectively reduces the errors of motion prediction compared with previous methods,especially of long-term prediction.
基金partly supported by National Natural Science Foundation of China (Grant No. 11272209)National Basic Research Program of China (Grant No. 2011CB309704)State Key Laboratory of Ocean Engineering of China (Grant No. GKZD010056).
文摘Using 1200 CPUs of the National Supercomputer TH-A1 and a parallel integral algorithm based on the 3500th-order Taylor expansion and the 4180-digit multiple precision data,we have done a reliable simulation of chaotic solution of Lorenz equation in a rather long interval 0 t 10000 LTU(Lorenz time unit).Such a kind of mathematically reliable chaotic simulation has never been reported.It provides us a numerical benchmark for mathematically reliable long-term prediction of chaos.Besides,it also proposes a safe method for mathematically reliable simulations of chaos in a finite but long enough interval.In addition,our very fine simulations suggest that such a kind of mathematically reliable long-term prediction of chaotic solution might have no physical meanings,because the inherent physical micro-level uncertainty due to thermal fluctuation might quickly transfer into macroscopic uncertainty so that trajectories for a long enough time would be essentially uncertain in physics.
基金Project supported by the National Natural Science Foundation of China(No.61371162)
文摘The latest video coding standard High Efficiency Video Coding （HEVC） can achieve much higher coding efficiency than previous video coding standards. Particularly, by exploiting the hierarchical B-picture prediction structure, temporal redundancy among neighbor frarnes is eliminated remarkably well. In practice, videos available to consumers usually contain many repeated shots, such as TV series, movies, and talk shows. According to our observations, when these videos are encoded by HEVC with the hierarchical B-picture structure, the temporal correlation in each shot is well exploited. However, the long-term correlation between repeated shots has not been used. We propose a long-term prediction （LTP） scheme to use the long-term temporal correlation between correlated shots in a video. The long-term reference （LTR） frames of a source video are chosen by clustering similar shots and extracting the representative frames, and a modified hierarchical B-picture coding structure based on an LTR frame is introduced to support long-term temporal prediction. An adaptive quantization method is further designed for LTR frames to improve the overall video coding efficiency. Experimental results show that up to 22.86% coding gain can be achieved using the new coding scheme.
