arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2005.05659cs.CV
收藏

Stillleben:机器人中用于深度学习的真实场景合成

Stillleben: Realistic Scene Synthesis for Deep Learning in Robotics

University of Bonn(波恩大学)

Max Schwarz, Sven Behnke

摘要:

训练数据是深度学习方法的关键成分,但对于机器人技术中经常遇到的专业领域来说,很难获得。我们描述了一个合成流水线,能够产生训练数据的杂乱的场景感知任务,如语义分割,目标检测,和对应或姿态估计。我们的方法安排对象网格在物理上逼真的,密集的场景,使用物理模拟。排列的场景使用高质量的光栅化和随机的外观和材质参数来渲染。模拟由相机传感器引入的噪声和其他变换。我们的管道可以在深度神经网络的训练过程中在线运行,从而在终身学习和迭代渲染和比较方法中产生应用。我们通过在具有挑战性的YCB视频数据集上学习语义分割来展示可用性,而实际上不使用任何训练帧,其中我们的方法实现了与传统训练模型相当的性能。此外,我们成功地应用在一个现实世界的regrasping系统。

英文摘要:

Training data is the key ingredient for deep learning approaches, but difficult to obtain for the specialized domains often encountered in robotics. We describe a synthesis pipeline capable of producing training data for cluttered scene perception tasks such as semantic segmentation, object detection, and correspondence or pose estimation. Our approach arranges object meshes in physically realistic, dense scenes using physics simulation. The arranged scenes are rendered using high-quality rasterization with randomized appearance and material parameters. Noise and other transformations introduced by the camera sensors are simulated. Our pipeline can be run online during training of a deep neural network, yielding applications in life-long learning and in iterative render-and-compare approaches. We demonstrate the usability by learning semantic segmentation on the challenging YCB-Video dataset without actually using any training frames, where our method achieves performance comparable to a conventionally trained model. Additionally, we show successful application in a real-world regrasping system.

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。

↑