AICurious Logo

What is: Bottom-up Path Augmentation?

SourcePath Aggregation Network for Instance Segmentation
Year2000
Data SourceCC BY-SA - https://paperswithcode.com

Bottom-up Path Augmentation is a feature extraction technique that seeks to shorten the information path and enhance a feature pyramid with accurate localization signals existing in low-levels. This is based on the fact that high response to edges or instance parts is a strong indicator to accurately localize instances.

Each building block takes a higher resolution feature map N_iN\_{i} and a coarser map P_i+1P\_{i+1} through lateral connection and generates the new feature map N_i+1N\_{i+1} Each feature map N_iN\_{i} first goes through a 3×33 \times 3 convolutional layer with stride 22 to reduce the spatial size. Then each element of feature map P_i+1P\_{i+1} and the down-sampled map are added through lateral connection. The fused feature map is then processed by another 3×33 \times 3 convolutional layer to generate N_i+1N\_{i+1} for following sub-networks. This is an iterative process and terminates after approaching P_5P\_{5}. In these building blocks, we consistently use channel 256 of feature maps. The feature grid for each proposal is then pooled from new feature maps, i.e., {N_2N\_{2}, N_3N\_{3}, N_4N\_{4}, N_5N\_{5}}.