arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2102.09532cs.CVcs.AIcs.LG
收藏

发条式变分自编码器

Clockwork Variational Autoencoders

University of Toronto(多伦多大学);Google Research, Brain Team(谷歌研究院大脑团队)

Vaibhav Saxena, Jimmy Ba, Danijar Hafner

摘要:

深度学习使算法能够生成逼真的图像。然而,准确预测长视频序列需要了解长期依赖关系,仍然是一个开放的挑战。虽然现有的视频预测模型成功地生成了清晰的图像,但它们往往无法准确地预测未来。我们引入了Clockwork VAE(CW-VAE),这是一种视频预测模型,它利用了潜在序列的层次结构,其中较高的级别以较慢的时间间隔进行。我们证明了分层潜伏期和时间抽象的好处4不同的视频预测数据集的序列高达1000帧,其中CW-VAE优于顶级视频预测模型。此外,我们还提出了一个Minecraft长期视频预测基准。我们进行了几个实验来深入了解CW-VAE,并确认较慢的水平学习表示视频中变化较慢的对象,而较快的水平学习表示较快的对象。

英文摘要:

Deep learning has enabled algorithms to generate realistic images. However, accurately predicting long video sequences requires understanding long-term dependencies and remains an open challenge. While existing video prediction models succeed at generating sharp images, they tend to fail at accurately predicting far into the future. We introduce the Clockwork VAE (CW-VAE), a video prediction model that leverages a hierarchy of latent sequences, where higher levels tick at slower intervals. We demonstrate the benefits of both hierarchical latents and temporal abstraction on 4 diverse video prediction datasets with sequences of up to 1000 frames, where CW-VAE outperforms top video prediction models. Additionally, we propose a Minecraft benchmark for long-term video prediction. We conduct several experiments to gain insights into CW-VAE and confirm that slower levels learn to represent objects that change more slowly in the video, and faster levels learn to represent faster objects.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑