Saliency based multiple object cosegmentation by ensemble MIML learning
- PDF / 4,600,043 Bytes
- 30 Pages / 439.642 x 666.49 pts Page_size
- 14 Downloads / 218 Views
Saliency based multiple object cosegmentation by ensemble MIML learning Bo Li1,2 · Zhengxing Sun1
· Junfeng Xu1 · Shuang Wang1 · Peiwen Yu1
Received: 3 September 2019 / Revised: 20 July 2020 / Accepted: 28 July 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract As an interesting and emerging topic, multiple foreground cosegmentation (MFC) aims at extracting a finite number of common objects from an image collection, which is useful to variety of visual media applications. Although a number of approaches have been proposed to address this problem, many of them are designed with the misleading consistent information, suboptimal image representation, or inefficient segmentation assist and thus still suffer from certain limitations, which reduces their capability in the real-world scenarios. To alleviate these limitations, we propose a novel unsupervised MFC framework, which is composed of three components: unsupervised label generation, saliency based pseudoannotation and cosegmentation by MIML learning. Specifically, we combine the high-level and low-level feature to represent the proposal objects, and adopt a novel SPAP clustering scheme to obtain more accurate consistent information of common objects. Then the saliency based pseudo-annotation help us reformulate the MFC problem as a Multi-Instance Multi-Label (MIML) learning problem by label propagation. Finally, by introducing a novel ensemble MIML learning scheme, the consistent information of common objects can more efficiently assist the segmentation of the images and get the more accurate segmentation results. We evaluate our framework on widely used public databases including the ICoseg dataset, MSRC dataset and FlickrMFC dataset for single and multiple common object cosegmentation respectively. Comparison results show that the proposed methods reach advanced and efficient performance. Keywords Multiple foreground cosegmentation · Saliency pseudo-annotation · Label propagation · Ensemble MIML learning
1 introduction Cosegmentation, proposed by Rother et al. [77], aims at jointly extracting common objects from an image collection, which makes full use of consistent constraint shared by all images Zhengxing Sun
[email protected]
Extended author information available on the last page of the article.
Multimedia Tools and Applications
in the collection to assist to segment every single image at the same time. A large number of cosegmentation methods have been proposed [42, 45, 46, 48, 74], and they confirm that results of cosegmentation are much better than results of classical single image segmentation. Nevertheless, these methods require that every single image in the image collection contain the same common objects. Kim et al. [47] address a challenging image cosegmentation problem called multiple foreground cosegmentation , i.e. MFC. The aim of MFC is to extract a finite number of common objects from an image collection, while only an unknown subset of common objects is presented in every single image. MFC relaxes the
Data Loading...
