Recently, learned video compression has drawn lots of attention and show a rapid development trend with promising results. However, the previous works still suffer from some criticial issues and have a performance gap with traditional compression standards in terms of widely used PSNR metric. In this paper, we propose several techniques to effectively improve the performance. First, to address the problem of accumulative error, we introduce a conditional-I-frame as the first frame in the GoP, which stabilizes the reconstructed quality and saves the bit-rate. Second, to efficiently improve the accuracy of inter prediction without increasing the complexity of decoder, we propose a pixel-to-feature motion prediction method at encoder side that helps us to obtain high-quality motion information. Third, we propose a probability-based entropy skipping method, which not only brings performance gain, but also greatly reduces the runtime of entropy coding. With these powerful techniques, this paper proposes AlphaVC, a high-performance and efficient learned video compression scheme. To the best of our knowledge, AlphaVC is the first E2E AI codec that exceeds the latest compression standard VVC on all common test datasets for both PSNR (-28.2% BD-rate saving) and MSSSIM (-52.2% BD-rate saving), and has very fast encoding (0.001x VVC) and decoding (1.69x VVC) speeds.
翻译:最近,所学的视频压缩引起了人们的极大关注,并展示了令人充满希望的结果的快速发展趋势。然而,先前的作品仍然受到一些批评问题的影响,并且从广泛使用的 PSNR 度量来看,与传统的压缩标准存在绩效差距。在本文中,我们提出了有效改进性能的几种技术。首先,为解决累积错误问题,我们引入了一个有条件的一框架作为GoP的第一个框架,该框架稳定了重建的质量并节省了比特率。第二,为了在不增加解码器复杂性的情况下有效地提高相互预测的准确性,我们提议在编码器一侧采用像素到速度的动作预测方法,帮助我们获得高质量的运动信息。第三,我们提出了一种基于概率的英特普跳法方法,这不仅能增益,而且还大大缩短了加密编码的运行时间。有了这些强大的技术,本文提出了一种高性能和高效的视频压缩计划。对于我们的知识来说,阿尔法VC是第一个E2E AI2.2D 动作预测方法,它帮助我们获取高质量的运动信息。第三,我们提出了一种基于最新标准VC标准的VC 和MS-C 保存率的Pral-ral press press press press b-ral press press press press press press press press press b2% press b-pressal pressal pressional press b2. press b-pressal press press press b-press press press pressal press press press pressal pressal pral pral pral press press press press b2. bal press b2.% bal 和MS2x%%% MS2x% MS2x MS 2% MS2xal MS2x bal_VC)