Federated Online Adaptation for Deep Stereo
This work addresses the challenge of improving stereo network accuracy for devices with limited resources, though it is incremental as it builds on existing federated learning principles.
The paper tackles the problem of adapting deep stereo networks in resource-constrained environments by introducing a federated learning framework that enables collaborative optimization across multiple clients, achieving accuracy equivalent to on-device adaptation and better performance in challenging scenarios.
We introduce a novel approach for adapting deep stereo networks in a collaborative manner. By building over principles of federated learning, we develop a distributed framework allowing for demanding the optimization process to a number of clients deployed in different environments. This makes it possible, for a deep stereo network running on resourced-constrained devices, to capitalize on the adaptation process carried out by other instances of the same architecture, and thus improve its accuracy in challenging environments even when it cannot carry out adaptation on its own. Experimental results show how federated adaptation performs equivalently to on-device adaptation, and even better when dealing with challenging environments.