Interactive Object Segmentation With Inside-Outside Guidance

This paper explores how to harvest precise object segmentation masks while minimizing the human interaction cost. To achieve this, we propose an Inside-Outside Guidance (IOG) approach in this work... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK BENCHMARK
Interactive Segmentation Cityscapes val IOG Instance Average IoU 83.8 # 1
Interactive Segmentation COCO IOG Instance Average IoU 85.2 # 1
Interactive Segmentation PASCAL2COCO(Unseen) IOG Instance Average IoU 82.1 # 1
Interactive Segmentation Rooftop IOG Instance Average IoU 94.0 # 1
Interactive Segmentation ssTEM IOG Instance Average IoU 83.7 # 1

Methods used in the Paper


METHOD TYPE
1x1 Convolution
Convolutions
Residual Connection
Skip Connections
Bottleneck Residual Block
Skip Connection Blocks
Max Pooling
Pooling Operations
Residual Block
Skip Connection Blocks
Kaiming Initialization
Initialization
Global Average Pooling
Pooling Operations
ResNet
Convolutional Neural Networks
ReLU
Activation Functions
Batch Normalization
Normalization
Convolution
Convolutions
Average Pooling
Pooling Operations
Dilated Convolution
Convolutions
Pyramid Pooling Module
Semantic Segmentation Modules
DEXTR
Image Segmentation Models