Explore Intrinsic Geometry for Query-based Tiny and Oriented Object Detector with Momentum-based Bipartite Matching

TL;DR

This paper introduces IGOFormer, a novel query-based oriented object detector that incorporates intrinsic geometry and momentum-based matching to improve tiny and arbitrarily oriented object detection, especially in aerial imagery.

Contribution

The paper proposes a new detector that explicitly models intrinsic geometry and stabilizes inter-stage matching, enhancing performance on tiny, oriented objects.

Findings

Achieves 78.00% AP50 on DOTA-V1.0 with Swin-T backbone.

Outperforms existing methods in aerial oriented object detection.

Demonstrates the effectiveness of geometric embeddings and momentum-based matching.

Abstract

Recent query-based detectors have achieved remarkable progress, yet their performance remains constrained when handling objects with arbitrary orientations, especially for tiny objects capturing limited texture information. This limitation primarily stems from the underutilization of intrinsic geometry during pixel-based feature decoding and the occurrence of inter-stage matching inconsistency caused by stage-wise bipartite matching. To tackle these challenges, we present IGOFormer, a novel query-based oriented object detector that explicitly integrates intrinsic geometry into feature decoding and enhances inter-stage matching stability. Specifically, we design an Intrinsic Geometry-aware Decoder, which enhances the object-related features conditioned on an object query by injecting complementary geometric embeddings extrapolated from their correlations to capture the geometric layout of the object, thereby offering a critical geometric insight into its orientation. Meanwhile, a Momentum-based Bipartite Matching scheme is developed to adaptively aggregate historical matching costs by formulating an exponential moving average with query-specific smoothing factors, effectively preventing conflicting supervisory signals arising from inter-stage matching inconsistency. Extensive experiments and ablation studies demonstrate the superiority of our IGOFormer for aerial oriented object detection, achieving an AP$_{50}$ score of 78.00\% on DOTA-V1.0 using Swin-T backbone under the single-scale setting. The code will be made publicly available.