iCoseg: Interactive co-segmentation with intelligent scribble guidance

By Dhruv Batra, Adarsh Kowdle, Devi Parikh, Jiebo Luo, Tsuhan Chen. 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2010

513 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/aef84912576b251c7ef2a9668f7f15a3cc667c8a

How researchers see this paper

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

In Computer Vision and Pattern Recognition

Citations in the corpus

Object cosegmentation (2011, methodology): “In [4] the authors used the iCoseg dataset in an interactive cosegmentation framework and, to the best of our knowledge, this dataset was never used in a fully automatic setting.”
Predicting Sufficient Annotation Strength for Interactive Foreground Segmentation (2013, methodology): “to efficiently identify the foreground region [8, 16, 2, 19, 12, 7, 1].”
Multi-class cosegmentation (2012, methodology): “[5] on two standard benchmarks, MSRC-v2 and iCoseg [23].”
BiCoS: A Bi-level co-segmentation method for image classification (2011, methodology): “We note that other published co-segmentation methods either do not scale to even this number of images or require user supervision [5], thus making [19] the natural choice for the comparison.”
GraphLab: A Distributed Framework for Machine Learning in the Cloud (2011, background): “Previous cosegmentation methods [42] have focused on processing frames in isolation.”
Unsupervised object discovery and localization in the wild: Part-based matching with bottom-up region proposals (2015, background): “Since then, this approach has been improved in numerous ways [4, 6, 23, 55], and extended to handle more general cases [7, 25, 40, 54].”
Hierarchical Subquery Evaluation for Active Learning on a Graph (2014, methodology): “Active learning has been successfully applied to many different computer vision problems including tracking [32], image categorization [17], object detection [31], semantic segmentation [30], and image [1] and video segmentation [8], with both human and automatic oracles [18].”
Distributed GraphLab: A Framework for Machine Learning in the Cloud (2012, unknown): “ilar texture and color characteristics. The resulting segmentation (Fig. 7(a)) can be used in scene understanding and other computer vision and robotics applications. Previous co-segmentation methods [3] have focused on processing frames in isolation. Instead, we developed a joint co-segmentation al”
A Self-Paced Multiple-Instance Learning Framework for Co-Saliency Detection (2015, methodology): “However, they mainly have two-fold differences: 1) co-saliency detection only focuses on detecting the common salient objects while co-segmentation methods also tend to segment the similar but non-salient background regions [25, 26]; 2) co-segmentation usually needs semi- or interactivesupervision [”
Deep Feature Factorization For Concept Discovery (2018, methodology): “Dataset The iCoseg dataset [1] is a popular benchmark for co-segmentation methods.”
Enriching Visual Knowledge Bases via Object Discovery and Segmentation (2014, background): “Our work is also related to co-segmentation, where the task is to simultaneously segment visually similar objects from multiple images at the same time [29, 28, 3, 11, 13, 36].”
What's the Point: Semantic Segmentation with Point Supervision (2015, methodology): “[21]. A typical indoor scene contains 23 objects [12], raising the annotation cost to tens of minutes per image. Methods have been developed to reduce the annotation cost through effective interfaces [28, 20, 3, 21, 39], e.g., through requesting human feedback only as necessary [17]. Co-segmentation”
Segmentation Propagation in ImageNet (2012, unknown): “Our joint segmentation model can be seen as a generalization of both Grabcut [35] and Batra et al. [36].”
Salient object detection: A survey (2014, background): “ject detection models only need a single input image. Co-saliency detection is closely related to the concept of image co-segmentation that aims to segment similar objects from multiple images [130], [131]. As stated in [127], three major differences exist between co-saliency and cosegmentation. Fir”
Multiple Choice Learning: Learning to Produce Multiple Structured Outputs (2012, methodology): “We used the co-segmentation dataset, iCoseg, of Batra et al .”
Visual Saliency Detection Based on Multiscale Deep CNN Features (2016, methodology): “To save space, the performance on the SED [50] and ICOSEG [51] datasets is no longer reported since these datasets are not challenging and not widely used.”
Salient Object Detection: A Discriminative Regional Feature Integration Approach (2013, background): “edu/projects/touch-coseg data set [4], including 38 groups of totally 643 images.”