AdaGraph: Unifying Predictive and Continuous Domain Adaptation through Graphs
This work addresses the problem of visual domain adaptation for autonomous systems, presenting a novel method that unifies predictive and continuous adaptation, though it appears incremental in combining existing concepts.
The paper tackles predictive domain adaptation, where no target data is available, by introducing AdaGraph, the first deep architecture that uses a graph to leverage information from auxiliary domains, and also adapts to incoming target data at test time for continuous adaptation. Experiments on three benchmark databases demonstrate its effectiveness.
The ability to categorize is a cornerstone of visual intelligence, and a key functionality for artificial, autonomous visual machines. This problem will never be solved without algorithms able to adapt and generalize across visual domains. Within the context of domain adaptation and generalization, this paper focuses on the predictive domain adaptation scenario, namely the case where no target data are available and the system has to learn to generalize from annotated source images plus unlabeled samples with associated metadata from auxiliary domains. Our contributionis the first deep architecture that tackles predictive domainadaptation, able to leverage over the information broughtby the auxiliary domains through a graph. Moreover, we present a simple yet effective strategy that allows us to take advantage of the incoming target data at test time, in a continuous domain adaptation scenario. Experiments on three benchmark databases support the value of our approach.