Abstract: Training deep/convolutional neural networks (DNNs/CNNs) requires a large amount of memory and iterative computation, which necessitates speedup and energy reduction, especially for edge ...
To continue reading this content, please enable JavaScript in your browser settings and refresh this page. Preview this article 1 min The 13,600-square-foot space ...
Abstract: In this work we address the problem of 3D object detection from point clouds in data-limited environments. Training with simulated data is a common approach in such scenarios; however a ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results