IoU-uniform R-CNN: Breaking Through the Limitations of RPN

By Li Zhu, Zihao Xie, Liman Liu, Bo Tao, Wenbing Tao. 2019

58 citations recorded by the source corpus.

Region Proposal Network (RPN) is the cornerstone of two-stage object detectors, it generates a sparse set of object proposals and alleviates the extrem foregroundbackground class imbalance problem during training. However, we find that the potential of the detector has not been fully exploited due to the IoU distribution imbalance and inadequate quantity of the training samples generated by RPN. With the increasing intersection over union (IoU), the exponentially smaller numbers of positive samples would lead to the distribution skewed towards lower IoUs, which hinders the optimization of detector at high IoU levels. In this paper, to break through the limitations of RPN, we propose IoU-Uniform R-CNN, a simple but effective method that directly generates training samples with uniform IoU distribution for the regression branch as well as the IoU prediction branch. Besides, we improve the performance of IoU prediction branch by eliminating the feature offsets of RoIs at inference, which helps the NMS procedure by preserving accurately localized bounding box. Extensive experiments on the PASCAL VOC and MS COCO dataset show the effectiveness of our method, as well as its compatibility and adaptivity to many object detection architectures. The code is made publicly available at this https URL,

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Shuo Chen
2. Sinan Kalkan
3. Xiaolin Hu
4. Emre Akbas
5. Xiang-Yang Li
6. Baris Can Cam
7. Jun Yu Li
8. Lijun Wu

In Computer Vision and Pattern Recognition

1. Jian Yang
2. Wenhai Wang
3. Xiang-Yang Li
4. Luc Van Gool
5. Jinhui Tang
6. Shuo Chen
7. Sinan Kalkan
8. Xiaolin Hu

Citations in the corpus

• Generalized Focal Loss: Learning Qualified and Distributed Bounding Boxes for Dense Object Detection (2020, methodology): “the quality estimation is combined (usually multiplied) with classification confidence as final scores [12, 11, 26, 29, 35] f r the rank process of Non-Maximum Suppression (NMS) during inference.”
• Imbalance Problems in Object Detection: A Review (2019, methodology): “• IoU Loss [83] • Bounded IoU Loss [84] • GIoU Loss [85] • Distance IoU Loss [86] • Complete IoU Loss [86] • Cascade R-CNN [87] • HSD [88] • IoU-uniform R-CNN [89] • pRoI Generator [66]”