Made to Order: Discovering monotonic temporal changes via self-supervised video ordering
This addresses the challenge of identifying consistent changes over time in videos for applications in scene and object analysis, though it is incremental in its method adaptation.
The paper tackles the problem of discovering and localizing monotonic temporal changes in image sequences by using a self-supervised video ordering proxy task, achieving state-of-the-art results on standard benchmarks.
Our objective is to discover and localize monotonic temporal changes in a sequence of images. To achieve this, we exploit a simple proxy task of ordering a shuffled image sequence, with `time' serving as a supervisory signal, since only changes that are monotonic with time can give rise to the correct ordering. We also introduce a transformer-based model for ordering of image sequences of arbitrary length with built-in attribution maps. After training, the model successfully discovers and localizes monotonic changes while ignoring cyclic and stochastic ones. We demonstrate applications of the model in multiple domains covering different scene and object types, discovering both object-level and environmental changes in unseen sequences. We also demonstrate that the attention-based attribution maps function as effective prompts for segmenting the changing regions, and that the learned representations can be used for downstream applications. Finally, we show that the model achieves the state-of-the-art on standard benchmarks for image ordering.