We present a deep learning-based framework to synthesize motion in-betweening in a two-stage manner. Given some context frames and a target frame, the system can generate plausible transitions with variable lengths in a non-autoregressive fashion. The framework consists of two Transformer Encoder-based networks operating in two stages: in the first stage a Context Transformer is designed to generate rough transitions based on the context and in the second stage a Detail Transformer is employed to refine motion details. Compared to existing Transformer-based methods which either use a complete Transformer Encoder-Decoder architecture or additional 1D convolutions to generate motion transitions, our framework achieves superior performance with less trainable parameters by only leveraging the Transformer Encoder and masked self-attention mechanism. To enhance the generalization of our transformer-based framework, we further introduce Keyframe Positional Encoding and Learned Relative Positional Encoding to make our method robust in synthesizing longer transitions exceeding the maximum transition length during training. Our framework is also artist-friendly by supporting full and partial pose constraints within the transition, giving artists fine control over the synthesized results. We benchmark our framework on the LAFAN1 dataset, and experiments show that our method outperforms the current state-of-the-art methods at a large margin (an average of 16% for normal-length sequences and 55% for excessive-length sequences). Our method trains faster than the RNN-based method and achieves a four-time speedup during inference. We implement our framework into a production-ready tool inside an animation authoring software and conduct a pilot study to validate the practical value of our method.

No abstract

Background: Shanghai is one of the biggest cities which have the highest number of entry travelers from all over the world. The HIV(human immunodeficiency virus) infection status of this population can reflect global trends of HIV prevalence to a certain extent. Methods:A retrospective cohort study was conducted to reveal the prevalence and characteristics of HIV-1 infection among entry travelers who applied to residency in Shanghai. The HIV-1 infection rate was estimated based on the detection of HIV-1 antibody. Results:Among 50830 entry travelers who applied to residency in Shanghai (2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016), 245 were determined HIV-1 positive with an infection rate of 0.48%. The detection rate of HIV was significantly higher in male (P<0.0001). Those aged 18-30 years, 31-40 years and >40years accounted for 34.3%,39.6% and 26.1% respectively of the infected population. Although there was no trend of increase in HIV-1 prevalence rates (Cochran-Armitage Z =2.543, P =0.111),proportions of individuals infected through homosexual transmission increased over the study period (Cochran-Armitage Z =5.41, P<0.001), while the proportions infected through heterosexual(Cochran-Armitage Z=3.38, P=0.001). Conclusion: The rate and characteristics of HIV-1 infection among foreign applicant to residency in Shanghai were revealed in the study. The results could provide the necessary epidemiological data for monitoring the HIV-1 epidemic among entry international travelers and to further contribute to the establishment of relevant policies and regulations for HIV control and prevention.