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

By Yunchao Wei, Huaxin Xiao, Humphrey Shi, Zequn Jie, Jiashi Feng, Thomas S. Huang. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018

583 citations recorded by the source corpus.

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.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/f875489e265efcab0e0bb958248f3d49ae299c7f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Non-Salient Region Object Mining for Weakly Supervised Semantic Segmentation (2021, methodology): “[21], DCSP [8], TPL [27], GAIN [31], DSRG [23], MCOF [48], AffinityNet [2], RDC [52], SeeNet [22], OAA [24], ICD [17], BES [9], Fan et al.”
Semi-supervised Segmentation Based on Error-Correcting Supervision (2020, background): “Weakly-supervised segmentation models generate dense classiﬁcation maps despite only image level [32, 35] or bounding box annotations [17] being present.”
SDC – Stacked Dilated Convolution: A Unified Descriptor Network for Dense Matching Tasks (2019, methodology): “However, [8, 40] both exploit parallel dilated convolution for semantic context pooling, while we, for the ﬁrst time, use convolution with different dilation rates to compute multi-scale feature descriptors.”
Single-Stage Semantic Segmentation From Image Labels (2020, methodology): “We compute our final classification scores as yc ≡ ynGWP c + ysize-focal c and use the multi-label soft-margin loss function [39] used in previous work [3, 57] as the classification loss,”
Weakly-Supervised Salient Object Detection via Scribble Annotations (2020, methodology): “As discussed in [39], enlarging receptive fields by different dilation rates can propagate the discriminative information to non-discriminative object regions.”
Naive-Student: Leveraging Semi-Supervised Learning in Video Sequences for Urban Scene Segmentation (2020, methodology): “Semi-supervised learning has been widely applied to several computer vision tasks, including semantic segmentation [57,19,59,31,80,35,81,73,96], object detection [68,76,64], instance segmentation [35,60], panoptic segmentation [45], human pose estimation [58], person re-identification [93], multi-ob”
SG-One: Similarity Guidance Network for One-Shot Semantic Segmentation (2018, background): “image-level labels [25, 23, 24], scribbles [15, 21], bounding boxes [13, 6] and points [2].”
Railroad is not a Train: Saliency as Pseudo-pixel Supervision for Weakly Supervised Semantic Segmentation (2021, methodology): “Many WSSS methods [15, 20, 27, 28, 42, 44] exploit the saliency map as the background cues of pseudo-masks.”
Interactive Object Segmentation With Inside-Outside Guidance (2022, methodology): “Particularly, image-level labels [67], [68], [69], [70], [71], [72], points [73], [74], bounding boxes [75], [76], scribbles [77], [78], [79] have been employed as guidance to”
ADCrowdNet: An Attention-Injective Deformable Convolutional Network for Crowd Understanding (2018, background): “The multiple dilated convolution architecture is motivated from [34].”
Causal Intervention for Weakly-Supervised Semantic Segmentation (2020, methodology): “To this end, existing methods mainly focus on generating better seed areas [25, 57, 59, 58] and expanding these seed areas [1, 2, 20, 23, 55].”
Weakly Supervised Learning of Instance Segmentation With Inter-Pixel Relations (2019, unknown): “7 MDC [46] I Ground-truth Backgrounds 60.”
Survey on semantic segmentation using deep learning techniques (2019, methodology): “Wei et al. [133] presented a weakly and semi supervised approach by using multiple dilated convolutions.”
Weakly Supervised Scene Parsing with Point-based Distance Metric Learning (2018, background): “And (Wei et al. 2018) discuss the effect of dilated convolution in this task.”
FickleNet: Weakly and Semi-Supervised Semantic Image Segmentation Using Stochastic Inference (2019, methodology): “In addition, the receptive ﬁeld produced by a standard dilated convolution is square with a ﬁxed size, so that MDC tends to identify false positive regions.”
Weakly-Supervised Semantic Segmentation via Sub-Category Exploration (2020, methodology): “Numerous approaches [1, 12, 13, 18, 22, 38, 40] are proposed to refine the initial cue via expanding the region of attention map.”