Unsupervised Learning Methods for Visual Place Recognition in Discretely and Continuously Changing Environments
This addresses a practical problem for robotics and autonomous systems in dynamic environments, but it is incremental as it builds on existing methods with a new normalization approach.
The paper tackles the problem of visual place recognition when environmental conditions change within a sequence, showing that state-of-the-art CNN-based descriptors fail as same places under different conditions become less similar than different places under the same condition. It experimentally evaluates the impact of discrete and continuous changes, emphasizing the limitations of statistical standardization and proposing a novel clustering-based extension for unsupervised learning.
Visual place recognition in changing environments is the problem of finding matchings between two sets of observations, a query set and a reference set, despite severe appearance changes. Recently, image comparison using CNN-based descriptors showed very promising results. However, existing experiments from the literature typically assume a single distinctive condition within each set (e.g., reference: day, query: night). We demonstrate that as soon as the conditions change within one set (e.g., reference: day, query: traversal daytime-dusk-night-dawn), different places under the same condition can suddenly look more similar than same places under different conditions and state-of-the-art approaches like CNN-based descriptors fail. This paper discusses this practically very important problem of in-sequence condition changes and defines a hierarchy of problem setups from (1) no in-sequence changes, (2) discrete in-sequence changes, to (3) continuous in-sequence changes. We will experimentally evaluate the effect of these changes on two state-of-the-art CNN-descriptors. Our experiments emphasize the importance of statistical standardization of descriptors and shows its limitations in case of continuous changes. To address this practically most relevant setup, we investigate and experimentally evaluate the application of unsupervised learning methods using two available PCA-based approaches and propose a novel clustering-based extension of the statistical normalization.