360-Indoor: Towards Learning Real-World Objects in 360{\deg} Indoor Equirectangular Images

TL;DR

This paper introduces 360-Indoor, a comprehensive 360-degree indoor image dataset with extensive annotations, aiming to advance object detection and recognition in panoramic images for the vision community.

Contribution

The creation of the largest 360-degree indoor object detection dataset with detailed annotations and multiple object categories to facilitate research in panoramic vision.

Findings

State-of-the-art methods perform variably on 360-Indoor

The dataset enables benchmarking for detection and classification

Extensive annotations improve training and evaluation

Abstract

While there are several widely used object detection datasets, current computer vision algorithms are still limited in conventional images. Such images narrow our vision in a restricted region. On the other hand, 360{\deg} images provide a thorough sight. In this paper, our goal is to provide a standard dataset to facilitate the vision and machine learning communities in 360{\deg} domain. To facilitate the research, we present a real-world 360{\deg} panoramic object detection dataset, 360-Indoor, which is a new benchmark for visual object detection and class recognition in 360{\deg} indoor images. It is achieved by gathering images of complex indoor scenes containing common objects and the intensive annotated bounding field-of-view. In addition, 360-Indoor has several distinct properties: (1) the largest category number (37 labels in total). (2) the most complete annotations on average (27 bounding boxes per image). The selected 37 objects are all common in indoor scene. With around 3k images and 90k labels in total, 360-Indoor achieves the largest dataset for detection in 360{\deg} images. In the end, extensive experiments on the state-of-the-art methods for both classification and detection are provided. We will release this dataset in the near future.