Abstract:Learning pyramidal feature representations is crucial for recognizing object instances at different scales. Feature Pyramid Network (FPN) is the classic architecture to build a feature pyramid with high-level semantics throughout. However, intrinsic defects in feature extraction and fusion inhibit FPN from further aggregating more discriminative features. In this work, we propose Attention Aggregation based Feature Pyramid Network (A 2 -FPN), to improve multi-scale feature learning through attention-guided feature aggregation. In feature extraction, it extracts discriminative features by collecting-distributing multi-level global context features, and mitigates the semantic information loss due to drastically reduced channels. In feature fusion, it aggregates complementary information from adjacent features to generate location-wise reassembly kernels for content-aware sampling, and employs channel-wise reweighting to enhance the semantic consistency before element-wise addition. A 2 -FPN shows consistent gains on different instance segmentation frameworks. By replacing FPN with A 2 -FPN in Mask R-CNN, our model boosts the performance by 2.1% and 1.6% mask AP when using ResNet-50 and ResNet-101 as backbone, respectively. Moreover, A 2 -FPN achieves an improvement of 2.0% and 1.4% mask AP when integrated into the strong baselines such as Cascade Mask R-CNN and Hybrid Task Cascade.

CODH++: Macro-semantic differences oriented instance segmentation network

When Masked Image Modeling Meets Source-free Unsupervised Domain Adaptation: Dual-Level Masked Network for Semantic Segmentation

A<sup>2</sup>-FPN: Attention Aggregation based Feature Pyramid Network for Instance Segmentation

A Two-Pipeline Instance Segmentation Network via Boundary Enhancement for Scene Understanding

A^2-FPN: Attention Aggregation based Feature Pyramid Network for Instance Segmentation

EmbedMask: Embedding Coupling for One-stage Instance Segmentation

Mask R-Cnn With Feature Pyramid Attention For Instance Segmentation

Instance-aware Semantic Segmentation via Multi-task Network Cascades

Unifying Instance and Panoptic Segmentation with Dynamic Rank-1 Convolutions

Discriminative Features Reconstruction Network For Semantic Segmentation

Panoptic Feature Pyramid Networks

Mask Encoding: A General Instance Mask Representation for Object Segmentation

Attention Guided Global Enhancement and Local Refinement Network for Semantic Segmentation

MaskLab: Instance Segmentation by Refining Object Detection with Semantic and Direction Features

Enhanced Semantic Head for Cascade Instance Segmentation

FCPFNet: Feature Complementation Network with Pyramid Fusion for Semantic Segmentation

Partial Atrous Cascade R-CNN

Deep Object Co-segmentation via Spatial-Semantic Network Modulation

Cascaded panoptic segmentation method for high resolution remote sensing image

Improvement of Bounding Box and Instance Segmentation Accuracy Using ResNet-152 FPN with Modulated Deformable ConvNets v2 Backbone-based Mask Scoring R-CNN

Translation-aware Fully Convolutional Instance Segmentation