eduzhai > Applied Sciences > Engineering >

SIDOD A Synthetic Image Dataset for 3D Object Pose Recognition with Distractors

  • king
  • (0) Download
  • 20210507
  • Save

... pages left unread,continue reading

Document pages: 3 pages

Abstract: We present a new, publicly-available image dataset generated by the NVIDIADeep Learning Data Synthesizer intended for use in object detection, poseestimation, and tracking applications. This dataset contains 144k stereo imagepairs that synthetically combine 18 camera viewpoints of three photorealisticvirtual environments with up to 10 objects (chosen randomly from the 21 objectmodels of the YCB dataset [1]) and flying distractors. Object and camera pose,scene lighting, and quantity of objects and distractors were randomized. Eachprovided view includes RGB, depth, segmentation, and surface normal images, allpixel level. We describe our approach for domain randomization and provideinsight into the decisions that produced the dataset.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...