Deep learning based object detectors are commonly deployed on mobile devices to solve a
variety of tasks. For maximum accuracy, each detector is usually trained to solve one single
specific task, and comes with a completely independent set of parameters. While this
guarantees high performance, it is also highly inefficient, as each model has to be
separately downloaded and stored. In this paper we address the question: can task-specific
detectors be trained and represented as a shared set of weights, plus a very small set of …

As mobile hardware technology advances, on-device computation is becoming more and
more affordable. On one hand, efficient backbones like MobileNets optimize feature-
extraction costs by decomposing convolutions into more efficient operations. On the other
hand, one-stage detection approaches like SSD provide mobile-friendly detection heads. As
a result, detection models are now massively being moved from server-side to on-device.
While this constitutes great progress, it also brings new challenges. Specifically, multiple …