MessyTable: Instance Association in Multiple Camera Views

TL;DR

MessyTable introduces a complex multi-view dataset for instance association, highlighting the challenges of identifying and linking objects across views in cluttered, real-world scenes, and aims to inspire more robust solutions.

Contribution

This paper presents a new challenging dataset for multi-view instance association in cluttered scenes, revealing limitations of current methods and encouraging development of more robust algorithms.

Findings

Popular methods struggle with the dataset's complexity

Appearance differences and occlusions are key challenges

The dataset reveals gaps in current multi-view association techniques

Abstract

We present an interesting and challenging dataset that features a large number of scenes with messy tables captured from multiple camera views. Each scene in this dataset is highly complex, containing multiple object instances that could be identical, stacked and occluded by other instances. The key challenge is to associate all instances given the RGB image of all views. The seemingly simple task surprisingly fails many popular methods or heuristics that we assume good performance in object association. The dataset challenges existing methods in mining subtle appearance differences, reasoning based on contexts, and fusing appearance with geometric cues for establishing an association. We report interesting findings with some popular baselines, and discuss how this dataset could help inspire new problems and catalyse more robust formulations to tackle real-world instance association problems. Project page: $\href{https://caizhongang.github.io/projects/MessyTable/}{\text{MessyTable}}$