In this work we address the task of segmenting an object into its parts, or semantic part segmentation. We start by adapting a state-of-the-art semantic segmentation system to this task, and show that a combination of a fully-convolutional Deep CNN system coupled with Dense CRF labelling provides excellent results for a broad range of object categories. Still, this approach remains agnostic to high-level constraints between object parts. We introduce such prior information by means of the Restricted Boltzmann Machine, adapted to our task and train our model in an entirely discriminative fashion, as a hidden CRF, demonstrating that prior information can yield additional improvements. We evaluate the performance of our approach on the Penn-Fudan and LFW datasets for the tasks of pedestrian parsing and face labelling respectively. We show superior performance with respect to competitive methods that have been extensively engineered on these benchmarks, as well as realistic qualitative results on part segmentation, even in challenging cases, such as occluded or deformable objects. We also provide quantitative and extensive qualitative results on three classes from the PASCAL Parts dataset.

本研究探讨了物体分割的任务，通过使用语义分割系统和限制玻尔兹曼机的有益信息，结合Dense CRF标签，采用判别式方法提高了语义部分分割的性能，并在PASCAL数据集上展示了优越的性能。

基于高层引导的语义部分分割深度学习