arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.02069cs.CVcs.RO
收藏

用于6D姿态估计的Sim2Real实例级风格迁移

Sim2Real Instance-Level Style Transfer for 6D Pose Estimation

Woven Planet Holdings, Inc.(woven Planet控股公司)

Takuya Ikeda, Suomi Tanishige, Ayako Amma, Michael Sudano, Hervé Audren, Koichi Nishiwaki

AI总结:

针对6D姿态估计中合成与真实数据的域差距,提出无需人工干预的sim2real实例级风格迁移及完整训练流程,并在真实机器人平台上验证其显著提升。

摘要:

近年来,合成数据已被广泛用于训练6D姿态估计网络,部分原因在于它能够以低成本自动提供完美标注。然而,合成数据与真实数据之间仍存在不可忽视的域差距,例如纹理/材质方面的差异。这些差距会对性能产生可测量的影响。为解决这一问题,我们提出了一种用于6D姿态估计网络训练的从仿真到现实(sim2real)实例级风格迁移方法。我们的方法无需人工干预,即可将目标对象的风格从合成域单独迁移到真实域。这提高了用于训练姿态估计网络的合成数据质量。我们还提出了一条从数据采集到姿态估计网络训练的完整流程,并在真实机器人平台上进行了广泛评估。评估表明,我们的方法在姿态估计性能以及经风格迁移适配后的图像真实感方面均取得了显著提升。

英文摘要:

In recent years, synthetic data has been widely used in the training of 6D pose estimation networks, in part because it automatically provides perfect annotation at low cost. However, there are still non-trivial domain gaps, such as differences in textures/materials, between synthetic and real data. These gaps have a measurable impact on performance. To solve this problem, we introduce a simulation to reality (sim2real) instance-level style transfer for 6D pose estimation network training. Our approach transfers the style of target objects individually, from synthetic to real, without human intervention. This improves the quality of synthetic data for training pose estimation networks. We also propose a complete pipeline from data collection to the training of a pose estimation network and conduct extensive evaluation on a real-world robotic platform. Our evaluation shows significant improvement achieved by our method in both pose estimation performance and the realism of images adapted by the style transfer.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑