CR2PQ: Continuous Relative Rotary Positional Query for Dense Visual Representation Learning

Dense visual representation learning (DRL) shows promise for learning localized information in dense prediction tasks, but struggles with establishing pixel/patch correspondence across different views (cross-contrasting). Existing methods primarily rely on self-contrasting the same view with variations, limiting input variance and hindering downstream performance. This paper delves into the mechanisms of selfcontrasting and cross-contrasting, identifying the crux of the issue: transforming discrete positional embeddings to continuous representations. To address the correspondence problem, we propose a Continuous Relative Rotary Positional Query (CR2PQ), enabling patch-level representation learning. Our extensive experiments on standard datasets demonstrate state-of-the-art (SOTA) results. Compared to the previous SOTA method (PQCL), our approach achieves significant improvements on COCO: with 300 epochs of pretraining, CR2PQ obtains 3.4% mAP bb and 2.1% mAP mk improvements for detection and segmentation tasks, respectively. Furthermore, CR2PQ exhibits faster convergence, achieving 10.4% mAP bb and 7.9% mAP mk improvements over SOTA with just 40 epochs of pretraining.