Multiview RGB-D Dataset for Object Instance Detection

TL;DR

This paper introduces a comprehensive multi-view RGB-D dataset of kitchen scenes with annotations, along with a new multi-view 3D proposal method and recognition baselines, demonstrating increased challenge over existing datasets.

Contribution

The paper provides a new multi-view RGB-D dataset for kitchen scenes and proposes a novel 3D proposal generation method with recognition baselines.

Findings

Our dataset is more challenging than WRGB-D for object detection.

The multi-view proposal method improves detection accuracy.

Recognition baselines trained on synthetic data perform competitively.

Abstract

This paper presents a new multi-view RGB-D dataset of nine kitchen scenes, each containing several objects in realistic cluttered environments including a subset of objects from the BigBird dataset. The viewpoints of the scenes are densely sampled and objects in the scenes are annotated with bounding boxes and in the 3D point cloud. Also, an approach for detection and recognition is presented, which is comprised of two parts: i) a new multi-view 3D proposal generation method and ii) the development of several recognition baselines using AlexNet to score our proposals, which is trained either on crops of the dataset or on synthetically composited training images. Finally, we compare the performance of the object proposals and a detection baseline to the Washington RGB-D Scenes (WRGB-D) dataset and demonstrate that our Kitchen scenes dataset is more challenging for object detection and recognition. The dataset is available at: http://cs.gmu.edu/~robot/gmu-kitchens.html.