Revisiting Depth Completion from a Stereo Matching Perspective for Cross-domain Generalization
This work addresses domain-shift issues in depth completion for applications like autonomous driving, though it is incremental as it adapts existing stereo methods.
The paper tackles the problem of depth completion's poor cross-domain generalization by proposing a framework that uses stereo matching networks on virtually projected stereo pairs, achieving robust performance across different domains.
This paper proposes a new framework for depth completion robust against domain-shifting issues. It exploits the generalization capability of modern stereo networks to face depth completion, by processing fictitious stereo pairs obtained through a virtual pattern projection paradigm. Any stereo network or traditional stereo matcher can be seamlessly plugged into our framework, allowing for the deployment of a virtual stereo setup that is future-proof against advancement in the stereo field. Exhaustive experiments on cross-domain generalization support our claims. Hence, we argue that our framework can help depth completion to reach new deployment scenarios.