Cascade Mask R-CNN

Introduced by Cai et al. in Cascade R-CNN: Delving into High Quality Object Detection

Cascade Mask R-CNN extends Cascade R-CNN to instance segmentation, by adding a mask head to the cascade.

In the Mask R-CNN, the segmentation branch is inserted in parallel to the detection branch. However, the Cascade R-CNN has multiple detection branches. This raises the questions of 1) where to add the segmentation branch and 2) how many segmentation branches to add. The authors consider three strategies for mask prediction in the Cascade R-CNN. The first two strategies address the first question, adding a single mask prediction head at either the first or last stage of the Cascade R-CNN. Since the instances used to train the segmentation branch are the positives of the detection branch, their number varies in these two strategies. Placing the segmentation head later on the cascade leads to more examples. However, because segmentation is a pixel-wise operation, a large number of highly overlapping instances is not necessarily as helpful as for object detection, which is a patch-based operation. The third strategy addresses the second question, adding a segmentation branch to each cascade stage. This maximizes the diversity of samples used to learn the mask prediction task.

At inference time, all three strategies predict the segmentation masks on the patches produced by the final object detection stage, irrespective of the cascade stage on which the segmentation mask is implemented and how many segmentation branches there are.

Source: Cascade R-CNN: Delving into High Quality Object Detection

Latest Papers

CascadeTabNet: An approach for end to end table detection and structure recognition from image-based documents
| Devashish PrasadAyan GadpalKshitij KapadniManish VisaveKavita Sultanpure
CBNet: A Novel Composite Backbone Network Architecture for Object Detection
| Yudong LiuYongtao WangSiwei WangTingTing LiangQijie ZhaoZhi TangHaibin Ling
GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond
| Yue CaoJiarui XuStephen LinFangyun WeiHan Hu
Hybrid Task Cascade for Instance Segmentation
| Kai ChenJiangmiao PangJiaqi WangYu XiongXiaoxiao LiShuyang SunWansen FengZiwei LiuJianping ShiWanli OuyangChen Change LoyDahua Lin
Cascade R-CNN: Delving into High Quality Object Detection
| Zhaowei CaiNuno Vasconcelos


Object Detection 4 33.33%
Instance Segmentation 3 25.00%
Semantic Segmentation 2 16.67%
Image Augmentation 1 8.33%
Table Detection 1 8.33%
Object Recognition 1 8.33%


RoI Feature Extractors