A Dataset To Evaluate The Representations Learned By Video Prediction Models

Ryan Szeto, Simon Stent, German Ros, Jason J. Corso

Abstract: We present a parameterized synthetic dataset called Moving Symbols to support the objective study of video prediction networks. Using several instantiations of the dataset in which variation is explicitly controlled, we highlight issues in an existing state-of-the-art approach and propose the use of a performance metric with greater semantic meaning to improve experimental interpretability. Our dataset provides canonical test cases that will help the community better understand, and eventually improve, the representations learned by such networks in the future. Code is available at https://github.com/rszeto/moving-symbols.

TL;DR: We propose a new dataset to better understand the shortcomings of existing video prediction networks.

Keywords: video prediction, self-supervised learning, dataset, model interpretability