作者: Jia He , Chang-Su Kim , C.-C. Jay Kuo
DOI: 10.1007/978-981-4451-60-4_3
关键词:
摘要: Interactive image segmentation techniques are semiautomatic processing approaches. They used to track object boundaries and/or propagate labels other regions by following user guidance so that heterogeneous in one can be separated. User interactions provide the high-level information indicating “object” and “background” regions. Then, various features such as locations, color intensities, local gradients extracted separate desired objects from background. We introduce several interactive methods according different models features.