Skip to content

Latest commit

 

History

History

condinst

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
 
 
 
 
 
 

CondInst

CondInst: Conditional Convolutions for Instance Segmentation

Abstract

We propose a simple yet effective instance segmentation framework, termed CondInst (conditional convolutions for instance segmentation). Top-performing instance segmentation methods such as Mask R-CNN rely on ROI operations (typically ROIPool or ROIAlign) to obtain the final instance masks. In contrast, we propose to solve instance segmentation from a new perspective. Instead of using instancewise ROIs as inputs to a network of fixed weights, we employ dynamic instance-aware networks, conditioned on instances. CondInst enjoys two advantages: 1) Instance segmentation is solved by a fully convolutional network, eliminating the need for ROI cropping and feature alignment. 2) Due to the much improved capacity of dynamically-generated conditional convolutions, the mask head can be very compact (e.g., 3 conv. layers, each having only 8 channels), leading to significantly faster inference. We demonstrate a simpler instance segmentation method that can achieve improved performance in both accuracy and inference speed. On the COCO dataset, we outperform a few recent methods including welltuned Mask R-CNN baselines, without longer training schedules needed.

Results and Models

Backbone Style MS train Lr schd bbox AP mask AP Config Download
R-50 pytorch Y 1x 39.8 36.0 config model | log

Citation

@inproceedings{tian2020conditional,
  title     =  {Conditional Convolutions for Instance Segmentation},
  author    =  {Tian, Zhi and Shen, Chunhua and Chen, Hao},
  booktitle =  {Proc. Eur. Conf. Computer Vision (ECCV)},
  year      =  {2020}
}