计算机科学
人工智能
分割
任务(项目管理)
深层神经网络
采样(信号处理)
可微函数
深度学习
代表(政治)
模式识别(心理学)
图像(数学)
图像分割
人工神经网络
计算机视觉
机器学习
数学
工程类
数学分析
系统工程
滤波器(信号处理)
政治
法学
政治学
作者
Varun Jampani,Daofeng Sun,Mingyu Li,Ming–Hsuan Yang,Jan Kautz
出处
期刊:Cornell University - arXiv
日期:2018-07-26
标识
DOI:10.48550/arxiv.1807.10174
摘要
Superpixels provide an efficient low/mid-level representation of image data, which greatly reduces the number of image primitives for subsequent vision tasks. Existing superpixel algorithms are not differentiable, making them difficult to integrate into otherwise end-to-end trainable deep neural networks. We develop a new differentiable model for superpixel sampling that leverages deep networks for learning superpixel segmentation. The resulting "Superpixel Sampling Network" (SSN) is end-to-end trainable, which allows learning task-specific superpixels with flexible loss functions and has fast runtime. Extensive experimental analysis indicates that SSNs not only outperform existing superpixel algorithms on traditional segmentation benchmarks, but can also learn superpixels for other tasks. In addition, SSNs can be easily integrated into downstream deep networks resulting in performance improvements.
科研通智能强力驱动
Strongly Powered by AbleSci AI