Multiple Class Segmentation Using A Unified Framework over Mean-Shift Patches

Yang, Lin; Meer, Peter; Foran, David J.

doi:10.1109/cvpr.2007.383229

Cited by 122 publications

(97 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We split the database in two halves for training and testing. Following the protocol of the other works [26,33,14,36], we ignored the void class during training and evaluation, and horse and mountain classes were also ignored due to insuffiencient amount of representative in the database. The reported results are the pixel-wise accuracy, that is the percentage of pixels correctly classified with respect to the total number of pixels in the segmentation ground truth.…”

Section: Methodsmentioning

confidence: 99%

“…Note that the results were not obtained using the same training and test datasets. Gould et al [14] used 40% images for training, 60% for test, Shotton et al [26], Verbeek and Triggs [33] split the dataset in two halves, Yang et al [36] reports the average results over a 5-fold cross validation. Our best method (merging multiple segmentation with mean operator) ranks second with for global pixel-wise accuracy with 76.1 behind Gould (76.5).…”

Section: Impact Of the Relaxation Labellingmentioning

confidence: 99%

See 1 more Smart Citation

Segmentation-based multi-class semantic object detection

Vieux

Benois-Pineau

Domenger

et al. 2010

Multimed Tools Appl

View full text Add to dashboard Cite

In this paper we study the problem of the detection of semantic objects from known categories in images. Unlike existing techniques which operate at the pixel or at a patch level for recognition, we propose to rely on the categorization of image segments. Recent work has highlighted that image segments provide a sound support for visual object class recognition. In this work, we use image segments as primitives to extract robust features and train detection models for a predefined set of categories. Several segmentation algorithms are benchmarked and their performances for segment recognition are compared. We then propose two methods for enhancing the segments classification, one based on the fusion of the classification results obtained with the different segmentations, the other one based on the optimization of the global labelling by correcting local ambiguities between neighbor segments. We use as a benchmark the Microsoft MSRC-21 image database and show that our method competes with the current state-of-the-art.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Impact Of the Relaxation Labellingmentioning

confidence: 99%

Segmentation-based multi-class semantic object detection

Vieux

Benois-Pineau

Domenger

et al. 2010

Multimed Tools Appl

View full text Add to dashboard Cite

show abstract

“…A number of recent works have reported encouraging results on this dataset: Shotton et al (TextonBoost) [27] introduced novel shape-texture features in a boosted framework, and used a CRF to integrate these with other cues; Verbeek and Triggs [28] explored combining spatial field models (like MRFs) with aspect-based models (like PLSA, LDA); Yang et al [30] combined texture, keypoint spatial co-occurrence and global shape into a meanshift framework to perform multi-class segmentation of images. In order to enable direct comparisons against TextonBoost, we duplicate Shotton et al's experimental methodology [27] and employ a random split of 45% for training, 10% for validation, and 45% for testing, while maintaining a similar distribution of classes.…”

Section: -Class Msrcmentioning

confidence: 99%

Learning class-specific affinities for image labelling

Batra

Sukthankar

Chen

2008

2008 IEEE Conference on Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

Spectral clustering and eigenvector-based methods have become increasingly popular in segmentation and recognition. Although the choice of the pairwise similarity metric (or affinities) greatly influences the quality of the results, this choice is typically specified outside the learning framework. In this paper, we present an algorithm to learn class-specific similarity functions. Mapping our problem in a Conditional Random Fields (CRF) framework enables us to pose the task of learning affinities as parameter learning in undirected graphical models. There are two significant advances over previous work. First, we learn the affinity between a pair of data-points as a function of a pairwise feature and (in contrast with previous approaches) the classes to which these two data-points were mapped, allowing us to work with a richer class of affinities. Second, our formulation provides a principled probabilistic interpretation for learning all of the parameters that define these affinities. Using ground truth segmentations and labellings for training, we learn the parameters with the greatest discriminative power (in an MLE sense) on the training data. We demonstrate the power of this learning algorithm in the setting of joint segmentation and recognition of object classes. Specifically, even with very simple appearance features, the proposed method achieves state-of-the-art performance on standard datasets.

show abstract

“…Semantic segmentation has attracted a lot of attention, but most works have focused on the fully supervised setting, where pixel labels are given for the training images [15,14,12,21,22,26]. The basic approach was formulated in [22], where a conditional random field (CRF) was defined over image pixels with unary potentials learnt by a boosted decision tree classifier over texture-layout filters.…”

Section: Related Workmentioning

confidence: 99%

“…This problem is very challenging because the method has to recover latent pixel labels from just presence labels, before it can generalize from the training set to test images. Recently there has been significant progress in fully supervised semantic segmentation [15,14,12,21,22,26], although the problem is still unsolved. The disadvantage of fully supervised techniques is the need for manually labeling pixels in the training set, which is time consuming and expensive.…”

Section: Introductionmentioning

confidence: 99%

Weakly supervised semantic segmentation with a multi-image model

Vezhnevets

Ferrari

Buhmann

2011

2011 International Conference on Computer Vision

142

195

View full text Add to dashboard Cite

We propose a novel method for weakly supervised semantic segmentation. Training images are labeled only by the classes they contain, not by their location in the image. On test images instead, the method predicts a class label for every pixel. Our main innovation is a multi-image model (MIM) -a graphical model for recovering the pixel labels of the training images. The model connects superpixels from all training images in a data-driven fashion, based on their appearance similarity. For generalizing to new test images we integrate them into MIM using a learned multiple kernel metric, instead of learning conventional classifiers on the recovered pixel labels. We also introduce an "objectness" potential, that helps separating objects (e.g. car, dog, human) from background classes (e.g. grass, sky, road). In experiments on the MSRC 21 dataset and the LabelMe subset of [18], our technique outperforms previous weakly supervised methods and achieves accuracy comparable with fully supervised methods.

show abstract

Multiple Class Segmentation Using A Unified Framework over Mean-Shift Patches

Cited by 122 publications

References 24 publications

Segmentation-based multi-class semantic object detection

Segmentation-based multi-class semantic object detection

Learning class-specific affinities for image labelling

Weakly supervised semantic segmentation with a multi-image model

Contact Info

Product

Resources

About