.png){kind=icon}
.png){kind=icon}
Return
Remote-Sensing Image Segmentation Based on Implicit 3-D Scene Representation
.png){kind=icon}
PRE
AI
DOI：10.1109/LGRS.2022.3227392.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Remote-sensing image segmentation, as a challenging but fundamental task, has drawn increasing attention in the remote-sensing field. Recent advances in deep learning have greatly boosted research on this task. However, the existing deep-learning-based segmentation methods heavily rely on a large amount of pixelwise labeled training data, and the labeling process is time-consuming and labor-intensive. In this letter, we focus on the scenario that leverages the 3-D structure of multiview images and a limited number of annotations to generate accurate novel view segmentation. Under this scenario, we propose a novel method for remote-sensing image segmentation based on implicit 3-D scene representation, which generates arbitrary-view segmentation output from limited segmentation annotations. The proposed method employs a two-stage training strategy. In the first stage, we optimize the implicit neural representations of a 3-D scene and encode their multiview images into a neural radiance field. In the second stage, we transform the scene color attribute into semantic labels and propose a ray-convolution network to aggregate local 3-D consistency cues across different locations. We also design a color-radiance network to help our method generalize to unseen views. Experiments on both synthetic and real-world data suggest that our method significantly outperforms deep convolutional neural networks (CNNs)-based methods and other view synthesis-based methods. We also show that the proposed method can be applied as a novel data augmentation approach that benefits CNN-based segmentation methods.
Keywords:
Image segmentation
implicit neural representations
neural radiance field
remote sensing
Journal
IF:
16.4
Papers:
1.0W
Citations:
5.1K