Weakly-Supervised Semantic Segmentation by Redistributing Region Scores Back to the Pixels

Krapac, Josip; Šegvić, Siniša

doi:10.1007/978-3-319-45886-1_31

Cited by 4 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In terms of segmentation quality, currently only methods based on deep convolutional networks [19,33] are strong enough to tackle segmentation datasets of difficulty similar to what fully-supervised methods can handle, such as the PASCAL VOC 2012 [9], which we make use of in this work. In particular, MIL-FCN [25], MIL-ILP [26] and the approaches of [4,18] leverage deep networks in a multiple instance learning setting, differing mainly in their pooling strategies, i.e. how they convert their internal spatial representation to per-image labels.…”

Section: Related Workmentioning

confidence: 99%

Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentation

Kolesnikov

Lampert

2016

Lecture Notes in Computer Science

637

708

View full text Add to dashboard Cite

Abstract. We introduce a new loss function for the weakly-supervised training of semantic image segmentation models based on three guiding principles: to seed with weak localization cues, to expand objects based on the information about which classes can occur in an image, and to constrain the segmentations to coincide with object boundaries. We show experimentally that training a deep convolutional neural network using the proposed loss function leads to substantially better segmentations than previous state-of-the-art methods on the challenging PASCAL VOC 2012 dataset. We furthermore give insight into the working mechanism of our method by a detailed experimental study that illustrates how the segmentation quality is affected by each term of the proposed loss function as well as their combinations.

show abstract

Section: Related Workmentioning

confidence: 99%

Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentation

Kolesnikov

Lampert

2016

Lecture Notes in Computer Science

637

708

View full text Add to dashboard Cite

show abstract

“…The mIoU values of each category on validation and testing datasets demonstrate the effectiveness of our method. We compare our method with some methods including SFR (Kim and Hwang, 2016), RSP (Krapac and Šegvić, 2016), CCNN, MIL-seg. The mIoU values of our method are the highest in most categories.…”

Section: Comparisons With Other Methodsmentioning

confidence: 99%

Saliency guided deep network for weakly-supervised image segmentation

Sun

2019

Pattern Recognition Letters

View full text Add to dashboard Cite

Weakly-supervised image segmentation is an important task in computer vision. A key problem is how to obtain high quality objects location from image-level category. Classification activation mapping is a common method which can be used to generate high-precise object location cues. However these location cues are generally very sparse and small such that they can not provide effective information for image segmentation. In this paper, we propose a saliency guided image segmentation network to resolve this problem. We employ a self-attention saliency method to generate subtle saliency maps, and render the location cues grow as seeds by seeded region growing method to expand pixel-level labels extent. In the process of seeds growing, we use the saliency values to weight the similarity between pixels to control the growing. Therefore saliency information could help generate discriminative object regions, and the effects of wrong salient pixels can be suppressed efficiently. Experimental results on a common segmentation dataset PASCAL VOC2012 demonstrate the effectiveness of our method.

show abstract

Markov random field for image synthesis with an application to traffic sign recognition

Haselhoff

Nunn

Müller

et al. 2017

2017 IEEE Intelligent Vehicles Symposium (IV)

View full text Add to dashboard Cite

Weakly-Supervised Semantic Segmentation by Redistributing Region Scores Back to the Pixels

Cited by 4 publications

References 19 publications

Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentation

Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentation

Saliency guided deep network for weakly-supervised image segmentation

Markov random field for image synthesis with an application to traffic sign recognition

Contact Info

Product

Resources

About