A Log-likelihood Regularized KL Divergence for Video Prediction with A 3D Convolutional Variational Recurrent Network

TL;DR

This paper presents a novel variational recurrent network with 3D convolutions and a log-likelihood regularized KL divergence, significantly improving video prediction quality and efficiency.

Contribution

It introduces a 3D convolutional variational recurrent model with a new regularization technique combining maximum likelihood and KL divergence.

Findings

Outperforms existing methods on multiple benchmarks

Requires fewer parameters than comparable models

Achieves higher prediction accuracy and generalization

Abstract

The use of latent variable models has shown to be a powerful tool for modeling probability distributions over sequences. In this paper, we introduce a new variational model that extends the recurrent network in two ways for the task of video frame prediction. First, we introduce 3D convolutions inside all modules including the recurrent model for future frame prediction, inputting and outputting a sequence of video frames at each timestep. This enables us to better exploit spatiotemporal information inside the variational recurrent model, allowing us to generate high-quality predictions. Second, we enhance the latent loss of the variational model by introducing a maximum likelihood estimate in addition to the KL divergence that is commonly used in variational models. This simple extension acts as a stronger regularizer in the variational autoencoder loss function and lets us obtain better results and generalizability. Experiments show that our model outperforms existing video prediction methods on several benchmarks while requiring fewer parameters.