YCB-Ev 1.1: Event-vision dataset for 6DoF object pose estimation

TL;DR

The YCB-Ev dataset offers synchronized RGB-D and event data with ground truth 6DoF object poses for 21 objects, enabling evaluation of pose estimation algorithms across modalities and datasets.

Contribution

This work introduces the first dataset with ground truth 6DoF poses for event streams, facilitating cross-modal and cross-dataset evaluation of pose estimation methods.

Findings

Evaluated two state-of-the-art algorithms on the new dataset.

Demonstrated the generalization of algorithms across datasets.

Provided a new benchmark for event-based pose estimation.

Abstract

Our work introduces the YCB-Ev dataset, which contains synchronized RGB-D frames and event data that enables evaluating 6DoF object pose estimation algorithms using these modalities. This dataset provides ground truth 6DoF object poses for the same 21 YCB objects that were used in the YCB-Video (YCB-V) dataset, allowing for cross-dataset algorithm performance evaluation. The dataset consists of 21 synchronized event and RGB-D sequences, totalling 13,851 frames (7 minutes and 43 seconds of event data). Notably, 12 of these sequences feature the same object arrangement as the YCB-V subset used in the BOP challenge. Ground truth poses are generated by detecting objects in the RGB-D frames, interpolating the poses to align with the event timestamps, and then transferring them to the event coordinate frame using extrinsic calibration. Our dataset is the first to provide ground truth 6DoF pose data for event streams. Furthermore, we evaluate the generalization capabilities of two state-of-the-art algorithms, which were pre-trained for the BOP challenge, using our novel YCB-V sequences. The dataset is publicly available at https://github.com/paroj/ycbev.