AAAI2020_results.zip

In tracking of time-varying low-rank models of time-varying matrices, our AAAI 2020 paper present a method robust to both uniformly-distributed measurement noise and arbitrarily-distributed "sparse" noise. In theory, we bound the tracking error. In practice, our use of randomised coordinate descent is scalable and allows for encouraging results on a benchmark (changedetection.net). This dataset are the outputs of our algorithm, as reported in the paper. <br>The source code for our paper is also available on-line at https://github.com/jmarecek/OnlineLowRank