Domain Contrast for Domain Adaptive Object Detection
It addresses domain shift in object detection for computer vision applications, presenting an incremental improvement with a novel loss function.
The paper tackles domain adaptive object detection by proposing Domain Contrast (DC), a plug-and-play method that improves transferability and discriminability, achieving significant performance gains over baselines and state-of-the-art on benchmarks.
We present Domain Contrast (DC), a simple yet effective approach inspired by contrastive learning for training domain adaptive detectors. DC is deduced from the error bound minimization perspective of a transferred model, and is implemented with cross-domain contrast loss which is plug-and-play. By minimizing cross-domain contrast loss, DC guarantees the transferability of detectors while naturally alleviating the class imbalance issue in the target domain. DC can be applied at either image level or region level, consistently improving detectors' transferability and discriminability. Extensive experiments on commonly used benchmarks show that DC improves the baseline and state-of-the-art by significant margins, while demonstrating great potential for large domain divergence.