Name
Abstract | ||
|---|---|---|
Image segmentation needs both local boundary position information and global object context information. The performance of the recent state-of-the-art method, fully convolutional networks, reaches a bottleneck due to the neural network limit after balancing between the two types of information simultaneously in an end-to-end training style. To overcome this problem, we divide the semantic image segmentation into temporal subtasks. First, we find a possible pixel position of some object boundary; then trace the boundary at steps within a limited length until the whole object is outlined. We present the first deep reinforcement learning approach to semantic image segmentation, called DeepOutline, which outperforms other algorithms in Coco detection leaderboard in the middle and large size person category in Coco val2017 dataset. Meanwhile, it provides an insight into a divide and conquer way by reinforcement learning on computer vision problems. |
Year | Venue | Field |
|---|---|---|
2018 | arXiv: Computer Vision and Pattern Recognition | Bottleneck,Pattern recognition,Computer science,Image segmentation,Semantic image segmentation,Pixel,Artificial intelligence,Divide and conquer algorithms,Artificial neural network,Reinforcement learning |
DocType | Volume | Citations |
Journal | abs/1804.04603 | 0 |
PageRank | References | Authors |
0.34 | 12 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Zhenxin Wang | 1 | 12 | 4.28 |
| Sayan Sarcar | 2 | 49 | 13.99 |
| Jingxin Liu | 3 | 0 | 2.70 |
| Yilin Zheng | 4 | 0 | 0.68 |
| Xiangshi Ren | 5 | 0 | 1.01 |