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 2003.06409cs.CVcs.LGcs.RO
收藏

视频场景理解的概率未来预测

Probabilistic Future Prediction for Video Scene Understanding

Wayve;University of Cambridge(剑桥大学)

Anthony Hu, Fergal Cotter, Nikhil Mohan, Corina Gurau, Alex Kendall

摘要:

我们提出了一种新的深度学习架构,用于从视频中进行概率未来预测。我们预测未来的语义,几何形状和复杂的现实世界的城市场景的运动,并使用这种表示来控制自动驾驶汽车。这项工作是第一个联合预测自我运动,静态场景,并以概率的方式,这使得采样一致,高度可能的未来从一个紧凑的潜在空间的动态代理的运动。我们的模型使用时空卷积模块从RGB视频中学习表示。除了作为学习的驾驶策略的输入之外,学习的表示还可以被显式地解码为未来的语义分割、深度和光流。为了模拟未来的随机性,我们引入了一种条件变分方法,该方法可以最大限度地减少当前分布(给定我们所看到的情况下可能发生的事情)和未来分布(我们所观察到的实际发生的事情)之间的差异。在推理过程中,通过从当前分布中采样来生成不同的未来。

英文摘要:

We present a novel deep learning architecture for probabilistic future prediction from video. We predict the future semantics, geometry and motion of complex real-world urban scenes and use this representation to control an autonomous vehicle. This work is the first to jointly predict ego-motion, static scene, and the motion of dynamic agents in a probabilistic manner, which allows sampling consistent, highly probable futures from a compact latent space. Our model learns a representation from RGB video with a spatio-temporal convolutional module. The learned representation can be explicitly decoded to future semantic segmentation, depth, and optical flow, in addition to being an input to a learnt driving policy. To model the stochasticity of the future, we introduce a conditional variational approach which minimises the divergence between the present distribution (what could happen given what we have seen) and the future distribution (what we observe actually happens). During inference, diverse futures are generated by sampling from the present distribution.

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。

↑