AINet: Association Implantation for Superpixel Segmentation

01/26/2021 ∙ by Yaxiong Wang, et al. ∙ 15

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.

READ FULL TEXT
POST COMMENT

Comments

dharini

Love the article. I would like to apply this method on my dataset. Would you pls suggest where I can find the code?

dharini

Love the article. I would like to apply your method and compare the results on my data set. I would really appreciate if you could share your GitHub repository (or codes) with me.

Authors

page 3

page 5

page 6

page 7

This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.