Towards Closing the Gap in Weakly Supervised Semantic Segmentation with DCNNs: Combining Local and Global Models

by   Christoph Mayer, et al.

Generating training sets for deep convolutional neural networks is a bottleneck for modern real-world applications. This is a demanding tasks for applications where annotating training data is costly, such as in semantic segmentation. In the literature, there is still a gap between the performance achieved by a network trained on full and on weak annotations. In this paper, we establish a strategy to measure this gap and to identify the ingredients necessary to close it. On scribbles, we establish state-of-the-art results comparable to the latest published ones (Tang et al., 2018, arXiv:1804.01346): we obtain a gap in mIoU of 2.4 with CRF post-processing (2.3 However, we use completely different ideas: combining local and global annotator models and regularising their prediction to train DeepLabV2. Finally, closing the gap was reported only recently for bounding boxes in Khoreva et al. (arXiv:1603.07485v2), by requiring 10x more training images. By simulating varying amounts of pixel-level annotations respecting scribble human annotations statistics, we show that our training strategy reacts to small increases in the amount of annotations and requires only 2-5x more annotated pixels, closing the gap with only 3.1 contributes new ideas towards closing the gap in real-world applications.


page 2

page 7

page 8

page 9

page 12


BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation

Recent leading approaches to semantic segmentation rely on deep convolut...

Tackling the Problem of Limited Data and Annotations in Semantic Segmentation

In this work, the case of semantic segmentation on a small image dataset...

Built-in Foreground/Background Prior for Weakly-Supervised Semantic Segmentation

Pixel-level annotations are expensive and time consuming to obtain. Henc...

COCO_TS Dataset: Pixel-level Annotations Based on Weak Supervision for Scene Text Segmentation

The absence of large scale datasets with pixel-level supervisions is a s...

A Centroid Loss for Weakly Supervised Semantic Segmentation in Quality Control and Inspection Application

Process automation has enabled a level of accuracy and productivity that...

Please sign up or login with your details

Forgot password? Click here to reset