Search papers, labs, and topics across Lattice.
This paper introduces CODE, a novel framework for Open World Object Detection (OWOD) that addresses semantic ambiguity in multimodal models by integrating Cross-Modal Joint Confidence Calibration, Uncertainty-Guided Universal Objectness Enhancement, and Dynamic Outlier Suppression. The method enhances known-class predictions with global visual prototypes, strengthens detection of unknown objects through uncertainty measures, and replaces rigid outlier penalties with a margin-aware approach. Experimental results on the Real-World Detection benchmark show that CODE significantly outperforms the previous state of the art, achieving 21.7 U-mAP and 40.8 K-mAP with the OWL-ViT L/14 backbone.
CODE outperforms prior methods in Open World Object Detection by effectively balancing known and unknown object detection through innovative calibration and suppression techniques.
Open World Object Detection (OWOD) built on multimodal foundation models often suffers from semantic ambiguity caused by unidirectional text-to-vision matching, while rigid outlier penalties may over-suppress unknown objects near known-class decision boundaries. We propose CODE (Cross-Modal Calibration and Dynamic Suppression), a unified inference-time framework with three complementary components. Cross-Modal Joint Confidence Calibration injects global visual prototypes to calibrate text-driven known-class predictions. Uncertainty-Guided Universal Objectness Enhancement measures classification hesitation from local visual responses to strengthen potential unknown objects. Dynamic Outlier Suppression via Confidence Margin replaces rigid suppression with a margin-aware adjustment that preserves ambiguous out-of-distribution instances. Experiments on the Real-World Detection benchmark demonstrate that, with the OWL-ViT L/14 backbone, CODE achieves 21.7 U-mAP and 40.8 K-mAP in Task 1, surpassing the previous state of the art by 2.6 and 2.3 points, respectively.
