Memorable Maps: A Framework for Re-defining Places in Visual Place Recognition
This work addresses visual place recognition for robotics and autonomous systems, offering an incremental improvement through a novel memorability framework.
The paper tackles the problem of visual place recognition by proposing a cognition-inspired framework that assesses image memorability using a tri-folded criterion based on human memorability, image entropy, and static content. The framework provides significant performance boosts to state-of-the-art methods, as demonstrated on the created ESSEX3IN1 dataset and other public datasets.
This paper presents a cognition-inspired agnostic framework for building a map for Visual Place Recognition. This framework draws inspiration from human-memorability, utilizes the traditional image entropy concept and computes the static content in an image; thereby presenting a tri-folded criterion to assess the 'memorability' of an image for visual place recognition. A dataset namely 'ESSEX3IN1' is created, composed of highly confusing images from indoor, outdoor and natural scenes for analysis. When used in conjunction with state-of-the-art visual place recognition methods, the proposed framework provides significant performance boost to these techniques, as evidenced by results on ESSEX3IN1 and other public datasets.