AINet: Association Implantation for Superpixel Segmentation

by   Yaxiong Wang, et al.

Recently, some approaches are proposed to harness deep convolutional networks to facilitate superpixel segmentation. The common practice is to first evenly divide the image into a pre-defined number of grids and then learn to associate each pixel with its surrounding grids. However, simply applying a series of convolution operations with limited receptive fields can only implicitly perceive the relations between the pixel and its surrounding grids. Consequently, existing methods often fail to provide an effective context when inferring the association map. To remedy this issue, we propose a novel Association Implantation (AI) module to enable the network to explicitly capture the relations between the pixel and its surrounding grids. The proposed AI module directly implants the features of grid cells to the surrounding of its corresponding central pixel, and conducts convolution on the padded window to adaptively transfer knowledge between them. With such an implantation operation, the network could explicitly harvest the pixel-grid level context, which is more in line with the target of superpixel segmentation comparing to the pixel-wise relation. Furthermore, to pursue better boundary precision, we design a boundary-perceiving loss to help the network discriminate the pixels around boundaries in hidden feature level, which could benefit the subsequent inferring modules to accurately identify more boundary pixels. Extensive experiments on BSDS500 and NYUv2 datasets show that our method could not only achieve state-of-the-art performance but maintain satisfactory inference efficiency.


page 3

page 5

page 6

page 7


Understanding Convolution for Semantic Segmentation

Recent advances in deep learning, especially deep convolutional neural n...

Multi Receptive Field Network for Semantic Segmentation

Semantic segmentation is one of the key tasks in computer vision, which ...

An Effective Pixel-Wise Approach for Skin Colour Segmentation Using Pixel Neighbourhood Technique

This paper presents a novel technique for skin colour segmentation that ...

CPP-Net: Context-aware Polygon Proposal Network for Nucleus Segmentation

Nucleus segmentation is a challenging task due to the crowded distributi...

AinnoSeg: Panoramic Segmentation with High Perfomance

Panoramic segmentation is a scene where image segmentation tasks is more...

Beyond Fixed Grid: Learning Geometric Image Representation with a Deformable Grid

In modern computer vision, images are typically represented as a fixed u...

Deep Neural Networks for Automatic Grain-matrix Segmentation in Plane and Cross-polarized Sandstone Photomicrographs

Grain segmentation of sandstone that is partitioning the grain from its ...