<p>A team of researchers at USC Viterbi&rsquo;s Thomas Lord Department of Computer Science has developed a new technique to copy virtual 3D objects and paste them into real indoor scenes. This creates an overall natural and realistic image in terms of spatial relationships, object orientations and lighting.&nbsp; The technique &ndash; called 3D Copy-Paste &ndash; can teach computers how to recognize the virtual 3D object in a multitude of different settings without having to rely on the tedious and expensive process of having a human feed the computer with reams of data.&nbsp;</p>

<p>The tool could have &ldquo;profound&rdquo; implications for both the computer graphics and computer vision fields, according to the researchers.&nbsp;Take, for example, autonomous driving technology.&nbsp;</p>

<p>An image of a cow is most associated with pastures and other bucolic settings.&nbsp;If you want to teach an AI in a self-driving car to avoid hitting a cow in front of your moving vehicle, the AI initially might get confused&mdash;a cow normally isn&rsquo;t found in the middle of a road. You would have to feed it an image of a cow in front of a car for it to recognize the object quickly.&nbsp;The&nbsp;tool allows a computer to recognize an object in an endless variety of environments without having to be frontloaded with a ton of images.&nbsp;</p>

Related Keywords

United States , America , Laurent Itti , Xinyu Huang , Yuliang Guo , Jiajun Wu , Cheng Zhao , Hong Xing Koven Yu , Liu Ren , Bosch Research North America , Viterbi Thomas Lord Department Of Computer Science , Stanford University , Thomas Lord Department , Computer Science , Physically Plausible Object Insertion , Neural Information Processing Systems , New Orleans , Assistant Professor ,