Statistical cues for domain specific image segmentation with performance analysis

By Scott Konishi, Alan L. Yuille. 2000

94 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology, Ocean Engineering

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

In Media Technology

In Ocean Engineering

Citations in the corpus

Per-Pixel Classification is Not All You Need for Semantic Segmentation (2021, background): “In early work, Konishi and Yuille [23] apply per-pixel Bayesian classifiers based on local image statistics.”
Geometric context from a single image (2005, neutral): “But in contrast to most recognition approaches that model semantic classes, such as cars, vegetation, roads, or buildings [21, 6, 14, 28], our goal is to model geometric classes that depend on the orientation of a physical object with relation to the scene.”
A Statistical Approach to Texture Classification from Single Images (2004, methodology): “Weak classification algorithms based on the statistical distribution of filter responses have been particularly successful of late (Cula and Dana, 2004; Konishi and Yuille, 2000; Leung and Malik, 2001; Schmid, 2001).”
Automatic photo pop-up (2005, background): “However, unlike most scene recognition approaches which aim to model semantic classes, such as cars, vegetation, roads, or buildings [Everingham et al. 1999; Konishi and Yuille 2000; Singhal et al. 2003], our goal is to model geometric classes that depend on the orientation of a physical object with”
A Statistical Approach to Texture Classification from Single Images (2005, methodology): “Weak classification algorithms based on the statistical distribution of filter responses have been particularly successful of late (Cula and Dana, 2004; Konishi and Yuille, 2000; Leung and Malik, 2001; Schmid, 2001).”
TextonBoost for Image Understanding: Multi-Class Object Recognition and Segmentation by Jointly Modeling Texture, Layout, and Context (2007, methodology): “Konishi and Yuille [26] labeled images using only a unary classifier, and hence did not achieve spatially coherent segmentations.”
Multiscale conditional random fields for image labeling (2004, background): “Note that the classifier’s performance is limited by class overlap and image noise [8].”
Classifying Images of Materials: Achieving Viewpoint and Illumination Independence (2002, methodology): “We would like to use the entire filter response of a texture for classification (in the manner of [ 12 ]),rather than its vector quantization into textons.”
Content based image retrieval using motif cooccurrence matrix (2004, methodology): “The filter response of the filter bank was learnt from training images and represented by clusters [34,35] or histograms [36,37].”
SAIL-VOS: Semantic Amodal Instance Level Video Object Segmentation – A Synthetic Dataset and Baselines (2019, methodology): “Classiﬁers for semantic segmentation were considered by Konishi and Yuille [57] demonstrating results on the Sowerby Image Database and the San Francisco Database.”
TextonBoost: Joint Appearance, Shape and Context Modeling for Multi-class Object Recognition and Segmentation (2006, background): “Konishi and Yuille [12] label images using a unary classifier and hence do not achieve spatially coherent segmentations.”
Constructing models for content-based image retrieval (2001, background): “Characteristic distributions of feat ure vectors are learnt by [ 8 , 14].”
A Statistical Approach to Material Classification Using Image Patch Exemplars (2009, methodology): “In this section, we describe the Columbia-Utrecht (CUReT) database [10] and its level of difﬁculty for single image classiﬁcation.”