Start Time: 14:10
End Time: 14:40
Many embedded vision applications require solutions that are robust in the face of very diverse real-world inputs. For example, in automotive applications, vision-based safety systems may encounter unusual configurations of road signs, or unfamiliar temporary barriers around construction sites. In this talk, we’ll present the approach that we use to address these “corner cases” in our development of Netradyne’s intelligent driver-safety monitoring system (IDMS). The essence of our approach is establishing a virtuous cycle which begins with running analytics at the edge and identifying scenarios of interest and corner cases on the embedded edge device. Data from these cases is then uploaded to the cloud, where it is labeled and then utilized for training new deep learning and analytics models. These new models are then deployed to the embedded device to enable improved performance and begin the cycle anew. We’ll also look at how we use this virtuous cycle to develop and deploy new features. Additionally, we’ll show how we leverage customer expertise to help identify corner cases at scale.