Learning Region Features for Object Detection
This addresses a key bottleneck in object detection for computer vision researchers, though it is incremental as it builds on existing methods.
The paper tackled the problem of hand-crafted region feature extraction in object detection by proposing a novel end-to-end learnable method that unifies existing approaches, outperforming RoI pooling counterparts and advancing towards fully learnable detection.
While most steps in the modern object detection methods are learnable, the region feature extraction step remains largely hand-crafted, featured by RoI pooling methods. This work proposes a general viewpoint that unifies existing region feature extraction methods and a novel method that is end-to-end learnable. The proposed method removes most heuristic choices and outperforms its RoI pooling counterparts. It moves further towards fully learnable object detection.