Cascade R-CNN: High Quality Object Detection and Instance Segmentation

by   Zhaowei Cai, et al.

In object detection, the intersection over union (IoU) threshold is frequently used to define positives/negatives. The threshold used to train a detector defines its quality. While the commonly used threshold of 0.5 leads to noisy (low-quality) detections, detection performance frequently degrades for larger thresholds. This paradox of high-quality detection has two causes: 1) overfitting, due to vanishing positive samples for large thresholds, and 2) inference-time quality mismatch between detector and test hypotheses. A multi-stage object detection architecture, the Cascade R-CNN, composed of a sequence of detectors trained with increasing IoU thresholds, is proposed to address these problems. The detectors are trained sequentially, using the output of a detector as training set for the next. This resampling progressively improves hypotheses quality, guaranteeing a positive training set of equivalent size for all detectors and minimizing overfitting. The same cascade is applied at inference, to eliminate quality mismatches between hypotheses and detectors. An implementation of the Cascade R-CNN without bells or whistles achieves state-of-the-art performance on the COCO dataset, and significantly improves high-quality detection on generic and specific object detection datasets, including VOC, KITTI, CityPerson, and WiderFace. Finally, the Cascade R-CNN is generalized to instance segmentation, with nontrivial improvements over the Mask R-CNN. To facilitate future research, two implementations are made available at <> (Caffe) and <> (Detectron).


page 1

page 7


Cascade R-CNN: Delving into High Quality Object Detection

In object detection, an intersection over union (IoU) threshold is requi...

Recursively Refined R-CNN: Instance Segmentation with Self-RoI Rebalancing

Within the field of instance segmentation, most of the state-of-the-art ...

Cascade-DETR: Delving into High-Quality Universal Object Detection

Object localization in general environments is a fundamental part of vis...

Rethinking Classification and Localization for Cascade R-CNN

We extend the state-of-the-art Cascade R-CNN with a simple feature shari...

Dynamic R-CNN: Towards High Quality Object Detection via Dynamic Training

Although two-stage object detectors have continuously advanced the state...

SCNet: Training Inference Sample Consistency for Instance Segmentation

Cascaded architectures have brought significant performance improvement ...

Cascade RPN: Delving into High-Quality Region Proposal Network with Adaptive Convolution

This paper considers an architecture referred to as Cascade Region Propo...

Please sign up or login with your details

Forgot password? Click here to reset