RoI Feature Extractors

Deformable RoI Pooling

Introduced by Dai et al. in Deformable Convolutional Networks

Deformable RoI Pooling adds an offset to each bin position in the regular bin partition of the RoI Pooling. Similarly, the offsets are learned from the preceding feature maps and the RoIs, enabling adaptive part localization for objects with different shapes.

Source: Deformable Convolutional Networks


Paper Code Results Date Stars


Task Papers Share
Object Detection 2 20.00%
Semantic Segmentation 2 20.00%
Action Detection 1 10.00%
Activity Detection 1 10.00%
Activity Recognition 1 10.00%
Autonomous Driving 1 10.00%
Decision Making 1 10.00%
Instance Segmentation 1 10.00%


Component Type
🤖 No Components Found You can add them if they exist; e.g. Mask R-CNN uses RoIAlign