# Simultaneous multi-view instance detection with learned geometric soft-constraints

## TL;DR

This paper introduces a learnable, end-to-end multi-view object detection method that jointly models geometry and appearance, validated on a new large-scale urban panorama dataset, outperforming baselines.

## Contribution

It provides a large-scale dataset, a custom annotation tool, and a novel joint learning approach for multi-view object detection and re-identification.

## Key findings

- Superior performance on urban panorama dataset
- Effective integration of geometric soft constraints
- Robust detection despite viewpoint and lighting changes

## Abstract

We propose to jointly learn multi-view geometry and warping between views of the same object instances for robust cross-view object detection. What makes multi-view object instance detection difficult are strong changes in viewpoint, lighting conditions, high similarity of neighbouring objects, and strong variability in scale. By turning object detection and instance re-identification in different views into a joint learning task, we are able to incorporate both image appearance and geometric soft constraints into a single, multi-view detection process that is learnable end-to-end. We validate our method on a new, large data set of street-level panoramas of urban objects and show superior performance compared to various baselines. Our contribution is threefold: a large-scale, publicly available data set for multi-view instance detection and re-identification; an annotation tool custom-tailored for multi-view instance detection; and a novel, holistic multi-view instance detection and re-identification method that jointly models geometry and appearance across views.