跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.19836 · v1

AGPF: Dynamic Source Selection for Feature Pyramid Fusion

首次提交 2026-11-15 14:33
版本更新 2026-11-15 14:33
雷达观测 2026-11-17 12:06
arXiv ↗PDF ↗

摘要

Feature pyramids provide dense predictors with features at several spatial resolutions. However, it remains largely unexplored which pyramid levels each output location should aggregate information from. We investigate this question by allowing each output location to attend to features from every pyramid level and measuring how attention is distributed across scales. Surprisingly, locations on the finest output grid allocate approximately 75\% of their attention to features two levels coarser, rather than to the current or adjacent levels. Even on the same output grid, the balance shifts toward coarser features as object size increases. These observations motivate Asymmetric Global Pyramid Fusion (AGPF), a neck that preserves the original output grids while allowing each location to dynamically select its contextual sources across scales. Each location interacts with a multiscale memory whose features are aligned in image coordinates, enabling flexible cross-scale aggregation without altering prediction resolutions. Compared with FPN, AGPF improves all 19 object detection and all seven instance segmentation configurations evaluated on COCO. A single fusion block improves Cascade Mask R-CNN by 1.7 box AP and 2.3 mask AP, and the gains consistently transfer to semantic and panoptic segmentation.

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-15 14:33 · 最新观测版本

AGPF: Dynamic Source Selection for Feature Pyramid Fusion

Feature pyramids provide dense predictors with features at several spatial resolutions. However, it remains largely unexplored which pyramid levels each output location should aggregate information from. We investigate this question by allowing each output location to attend to features from every pyramid level and measuring how attention is distributed across scales. Surprisingly, locations on the finest output grid allocate approximately 75\% of their attention to features two levels coarser, rather than to the current or adjacent levels. Even on the same output grid, the balance shifts toward coarser features as object size increases. These observations motivate Asymmetric Global Pyramid Fusion (AGPF), a neck that preserves the original output grids while allowing each location to dynamically select its contextual sources across scales. Each location interacts with a multiscale memory whose features are aligned in image coordinates, enabling flexible cross-scale aggregation without altering prediction resolutions. Compared with FPN, AGPF improves all 19 object detection and all seven instance segmentation configurations evaluated on COCO. A single fusion block improves Cascade Mask R-CNN by 1.7 box AP and 2.3 mask AP, and the gains consistently transfer to semantic and panoptic segmentation.

查看此版本解读 →查看此版本原文 ↗