Semantic Segmentation without Annotating Segments

Xia, Wei; Domokos, Csaba; Dong, Jian; Cheong, Loong-Fah; Yan, Shuicheng

doi:10.1109/iccv.2013.271

Cited by 42 publications

(46 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared with histogram or GMM fitting used by [16], they argue that the fore/background regions can be clustered better using this energy. Probably the most similar work to us is Xia [6], which proposes to generate the object shape by directly selecting the best overlapping segments that align well to the object boundary and thereafter integrate it into the subsequent graph-cut based inference algorithm to obtain the segmentation results. Segmentation performance of this method heavily relies on the shape based graph-cut process.…”

Section: Related Workmentioning

confidence: 99%

“…We utilize the generated object candidates as segment hypotheses bases. However, unlike [6], we operationalize this idea by exploring the usefulness of each segment towards object extraction based on an object reconstruction model. Furthermore, the pairwise correlation information of segment hypotheses can be preserved in our method, which is crucial to produce accurate and reliable results.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Object Extraction from Bounding Box Prior with Double Sparse Reconstruction

Dai

Ding

Yang

et al. 2015

2015 IEEE International Conference on Computer Vision Workshop (ICCVW)

View full text Add to dashboard Cite

Extracting objects from natural images has long been an active problem in image processing. Despite various attempts, it has not been completely solved up to date. Current state-of-the-art object proposal methods tend to extract a set of object segments from an image, and often these are consequential differences among these results for each image. Another type of methods strive to detect one object into a bounding box where some background parts are often covered. For these two methodologies, we observe: 1) there are generally some regions overlapped among different proposals, which are usually from one object; they could be as object 'segment hypotheses'; 2) pixels outside the detected bounding box could be as 'background hypotheses' as they are with high probability from the background. With them, we formulate the object extraction as a "double" sparse reconstruction problem in terms of the bounding box results. The idea is that object regions should be with small reconstruction errors to segment hypotheses bases, simultaneously, they should have large reconstruction errors to background hypotheses bases. Comprehensive experiments and evaluations on PASCAL VOC object segmentation dataset and GrabCut-50 database demonstrate the superiority of our built method. In particular, we achieve the state-of-theart performance for the object segmentation with bounding box prior on these two benchmark datasets.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Object Extraction from Bounding Box Prior with Double Sparse Reconstruction

Dai

Ding

Yang

et al. 2015

2015 IEEE International Conference on Computer Vision Workshop (ICCVW)

View full text Add to dashboard Cite

show abstract

“…Methods of this category, e.g., [1,3,12,15,18,20,24,27,29], aim to identify objects of interest and segment them out at the same time. Owing to the high flexibility in modeling the dependencies among variables and observations, CRFs have been widely adopted in the task.…”

Section: Related Workmentioning

confidence: 99%

“…Recent research efforts have been made on reducing the labeling cost for object segmentation. Weakly supervised methods or cosegmentation, e.g., [13,18,29], support training data labeled in the levels of images or bounding boxes, instead of object maps. As information regarding object classes has been annotated, the class-specific clues were extracted in these methods to enhance object segmentation.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Multiple Structured-Instance Learning for Semantic Segmentation with Uncertain Training Data

Chang

Lin

Hsu

2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

We present an approach MSIL-CRF that incorporates multiple instance learning (MIL) into conditional random fields (CRFs). It can generalize CRFs to work on training data with uncertain labels by the principle of MIL. In this work, it is applied to saving manual efforts on annotating training data for semantic segmentation. Specifically, we consider the setting in which the training dataset for semantic segmentation is a mixture of a few object segments and an abundant set of objects' bounding boxes. Our goal is to infer the unknown object segments enclosed by the bounding boxes so that they can serve as training data for semantic segmentation. To this end, we generate multiple segment hypotheses for each bounding box with the assumption that at least one hypothesis is close to the ground truth. By treating a bounding box as a bag with its segment hypotheses as structured instances, MSIL-CRF selects the most likely segment hypotheses by leveraging the knowledge derived from both the labeled and uncertain training data. The experimental results on the Pascal VOC segmentation task demonstrate that MSIL-CRF can provide effective alternatives to manually labeled segments for semantic segmentation.

show abstract

Towards Unified Object Detection and Semantic Segmentation

Dong¹,

Chen²,

Yan³

et al. 2014

Computer Vision – ECCV 2014

Self Cite

View full text Add to dashboard Cite

Object detection and semantic segmentation are two strongly correlated tasks, yet typically solved separately or sequentially with substantially different techniques. Motivated by the complementary effect observed from the typical failure cases of the two tasks, we propose a unified framework for joint object detection and semantic segmentation. By enforcing the consistency between final detection and segmentation results, our unified framework can effectively leverage the advantages of leading techniques for these two tasks. Furthermore, both local and global context information are integrated into the framework to better distinguish the ambiguous samples. By jointly optimizing the model parameters for all the components, the relative importance of different component is automatically learned for each category to guarantee the overall performance. Extensive experiments on the PASCAL VOC 2010 and 2012 datasets demonstrate encouraging performance of the proposed unified framework for both object detection and semantic segmentation tasks.

show abstract

Semantic Segmentation without Annotating Segments

Cited by 42 publications

References 17 publications

Object Extraction from Bounding Box Prior with Double Sparse Reconstruction

Object Extraction from Bounding Box Prior with Double Sparse Reconstruction

Multiple Structured-Instance Learning for Semantic Segmentation with Uncertain Training Data

Towards Unified Object Detection and Semantic Segmentation

Contact Info

Product

Resources

About