Private Multiparty Perception for Navigation
This addresses privacy concerns in multi-camera navigation systems for applications like autonomous navigation, though it appears incremental as it builds on existing multiparty perception methods.
The paper tackles the problem of navigating through cluttered environments using multiple cameras while preserving privacy, by learning a multiview scene representation that only supports navigation tasks and prevents inference of other information, achieving successful navigation in complex scenes as demonstrated on a new dataset.
We introduce a framework for navigating through cluttered environments by connecting multiple cameras together while simultaneously preserving privacy. Occlusions and obstacles in large environments are often challenging situations for navigation agents because the environment is not fully observable from a single camera view. Given multiple camera views of an environment, our approach learns to produce a multiview scene representation that can only be used for navigation, provably preventing one party from inferring anything beyond the output task. On a new navigation dataset that we will publicly release, experiments show that private multiparty representations allow navigation through complex scenes and around obstacles while jointly preserving privacy. Our approach scales to an arbitrary number of camera viewpoints. We believe developing visual representations that preserve privacy is increasingly important for many applications such as navigation.