Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentation

TL;DR

This paper introduces a simple yet effective method using varied dilation rates in convolutional networks to significantly improve weakly- and semi-supervised semantic segmentation, achieving new state-of-the-art results.

Contribution

It reveals how dilated convolution can be utilized to enhance object localization in weakly- and semi-supervised segmentation, a novel approach compared to prior methods.

Findings

Achieves 60.8% mIoU on Pascal VOC 2012 in weakly supervised setting

Achieves 67.6% mIoU in semi-supervised setting

Outperforms existing state-of-the-art methods

Abstract

Despite the remarkable progress, weakly supervised segmentation approaches are still inferior to their fully supervised counterparts. We obverse the performance gap mainly comes from their limitation on learning to produce high-quality dense object localization maps from image-level supervision. To mitigate such a gap, we revisit the dilated convolution [1] and reveal how it can be utilized in a novel way to effectively overcome this critical limitation of weakly supervised segmentation approaches. Specifically, we find that varying dilation rates can effectively enlarge the receptive fields of convolutional kernels and more importantly transfer the surrounding discriminative information to non-discriminative object regions, promoting the emergence of these regions in the object localization maps. Then, we design a generic classification network equipped with convolutional blocks of different dilated rates. It can produce dense and reliable object localization maps and effectively benefit both weakly- and semi- supervised semantic segmentation. Despite the apparent simplicity, our proposed approach obtains superior performance over state-of-the-arts. In particular, it achieves 60.8% and 67.6% mIoU scores on Pascal VOC 2012 test set in weakly- (only image-level labels are available) and semi- (1,464 segmentation masks are available) supervised settings, which are the new state-of-the-arts.