RFENet: Towards Reciprocal Feature Evolution for Glass Segmentation
This addresses a specific segmentation problem for computer vision applications, with incremental improvements in method design.
The paper tackles the problem of segmenting glass-like objects, which is challenging due to transparency and tiny boundaries, by proposing RFENet with modules for reciprocal feature learning and refinement, achieving state-of-the-art performance on three public datasets.
Glass-like objects are widespread in daily life but remain intractable to be segmented for most existing methods. The transparent property makes it difficult to be distinguished from background, while the tiny separation boundary further impedes the acquisition of their exact contour. In this paper, by revealing the key co-evolution demand of semantic and boundary learning, we propose a Selective Mutual Evolution (SME) module to enable the reciprocal feature learning between them. Then to exploit the global shape context, we propose a Structurally Attentive Refinement (SAR) module to conduct a fine-grained feature refinement for those ambiguous points around the boundary. Finally, to further utilize the multi-scale representation, we integrate the above two modules into a cascaded structure and then introduce a Reciprocal Feature Evolution Network (RFENet) for effective glass-like object segmentation. Extensive experiments demonstrate that our RFENet achieves state-of-the-art performance on three popular public datasets.