计算机科学
人工智能
均方误差
特征(语言学)
深度学习
光流
代表(政治)
像素
特征学习
功能(生物学)
序列(生物学)
卷积神经网络
帧(网络)
模式识别(心理学)
比例(比率)
图像(数学)
计算机视觉
机器学习
数学
统计
物理
政治
哲学
生物
进化生物学
电信
量子力学
遗传学
法学
语言学
政治学
作者
Michaël Mathieu,Camille Couprie,Yann LeCun
出处
期刊:International Conference on Learning Representations
日期:2016-01-01
被引量:1077
摘要
Abstract: Learning to predict future images from a video sequence involves the construction of an internal representation that models the image evolution accurately, and therefore, to some degree, its content and dynamics. This is why pixel-space video prediction may be viewed as a promising avenue for unsupervised feature learning. In addition, while optical flow has been a very studied problem in computer vision for a long time, future frame prediction is rarely approached. Still, many vision applications could benefit from the knowledge of the next frames of videos, that does not require the complexity of tracking every pixel trajectories. In this work, we train a convolutional network to generate future frames given an input sequence. To deal with the inherently blurry predictions obtained from the standard Mean Squared Error (MSE) loss function, we propose three different and complementary feature learning strategies: a multi-scale architecture, an adversarial training method, and an image gradient difference loss function. We compare our predictions to different published results based on recurrent neural networks on the UCF101 dataset
科研通智能强力驱动
Strongly Powered by AbleSci AI