Bal-R2CNN: High Quality Recurrent Object Detection With Balance Optimization

Abstract

It is a common practice to refine object detection results using recurrent detection paradigm. We evaluate the recurrent detection on Faster R-CNN, but the improvement is far away from expected. We consider that the performance bottleneck is from imbalance optimization caused by the biased distribution of training data. Low-IoU-skewed RPN proposals could suppress the contribution of High-IoU examples at the training stage. Besides, data imbalance and statistical discrepancy on regression targets between low-IoU and high-IoU examples are not considered in the regression task; this design could impede localization quality. In this work, we propose Bal-R2CNN for high-quality recurrent object detection. There are two new components in Bal-R2CNN. Self-iteration box sampling collects object boxes from recurrent steps and increases the number of high-IoU training examples. IoU-sensitive bounding-box regression sends proposal boxes with different IoUs to specified regression branches for more accurate bounding-box prediction. Both two new components could induce balanced optimization and be helpful. With the resulting Bal-R2CNN detector, evaluation on PASCAL VOC and MSCOCO reveal that our method has a significant improvement on the existing solution and could reach a better performance than several state-of-the-art methods.