Augmenting Proposals by the Detector Itself
This addresses a long-standing bottleneck in object detection for computer vision applications, offering a hyperparameter-free solution that integrates with advanced methods for incremental gains.
The paper tackles the problem of insufficient high-quality proposals in two-stage and multi-stage object detectors by introducing APDI, a novel training method that augments proposals using the detector itself, resulting in at least 2.7 AP improvements on Faster R-CNN with various backbones on the COCO dataset.
Lacking enough high quality proposals for RoI box head has impeded two-stage and multi-stage object detectors for a long time, and many previous works try to solve it via improving RPN's performance or manually generating proposals from ground truth. However, these methods either need huge training and inference costs or bring little improvements. In this paper, we design a novel training method named APDI, which means augmenting proposals by the detector itself and can generate proposals with higher quality. Furthermore, APDI makes it possible to integrate IoU head into RoI box head. And it does not add any hyperparameter, which is beneficial for future research and downstream tasks. Extensive experiments on COCO dataset show that our method brings at least 2.7 AP improvements on Faster R-CNN with various backbones, and APDI can cooperate with advanced RPNs, such as GA-RPN and Cascade RPN, to obtain extra gains. Furthermore, it brings significant improvements on Cascade R-CNN.