Graph Cut Based Inference with Co-occurrence Statistics

By Lubor Ladicky, Chris Russell, Pushmeet Kohli, Philip H. S. Torr. European Conference on Computer Vision, 2010

366 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/0095ca57665eae51c9dd7a4ed8a3311aeea1b441

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

A Sentence Is Worth a Thousand Pixels (2013, methodology): “level by averaging the TextonBoost [16] pixel potentials inside each region.”
Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials (2011, background): “This limits the ability of region-based approaches to produce accurate label assignments around complex object boundaries, although significant progress has been made [9, 13, 14].”
A High Performance CRF Model for Clothes Parsing (2014, methodology): “Co-occurrence potentials have been employed to enforce consistency among region labels [23].”
Scene Graph Generation by Iterative Message Passing (2017, background): “Visual scene understanding often harnesses the statistical patterns of object co-occurrence [11, 22, 30, 35] as well as spatial layout [2, 9].”
DeepContext: Context-Encoding Neural Pathways for 3D Holistic Scene Understanding (2016, background): “The role of context has been studied extensively in computer vision [27, 41, 5, 4, 3, 7, 18, 37, 33, 35, 34, 19, 12, 9, 43, 1, 39, 24, 10, 42, 20, 21].”
Context models and out-of-context objects (2012, methodology): “However, context models have been evaluated mostly based on the improvement of object recognition performance even though it is only one of many ways to exploit contextual information.”
Structured class-labels in random forests for semantic image labelling (2011, methodology): “For the task of semantic image labelling, state-of-the-art approaches [16, 17, 13, 24] are typically using complementary features at different levels within random field models [18].”
Figure-ground segmentation by transferring window masks (2012, methodology): “Figure-ground segmentation has recently been addressed successfully by interactive segmentation works [6, 26, 28, 35].”
Perceptual Organization and Recognition of Indoor Scenes from RGB-D Images (2013, background): “We observe that approaches like [4, 1], which focus on classifying bottom-up regions candidates using strong features on the region have obtained significantly better results than MRF-based methods [16].”
Visual Relationship Detection with Internal and External Linguistic Knowledge Distillation (2017, positive): “object cooccurrence, spatial relationships) to improve visual tasks [30, 27, 21, 14, 4, 5, 15].”
Relating Things and Stuff by High-Order Potential Modeling (2012, methodology): “We compare our full ACRF model with [1,20,21,12,13] on Stanford Background (refer as Stanford) dataset [1].”
Learning to segment under various forms of weak supervision (2015, methodology): “Semantic segmentation is one of the most fundamental challenges in computer vision, and conventional fully supervised algorithms have demonstrated promising results [21, 10, 8, 33, 35, 36, 44, 19, 20, 45, 31].”
Unrolling Loopy Top-Down Semantic Feedback in Convolutional Deep Networks (2014, methodology): “The first large group of methods is based on handcrafted appearance features coupled with a conditional random field (CRF) approach (of increasing complexity), which imposes spatial and semantic consistency in the final classification [15, 14, 16, 11, 31, 2].”
Building a Large-scale Multimodal Knowledge Base System for Answering Visual Queries (2015, background): “While MRFs have been widely used in a variety of vision tasks [9, 26, 27, 46], applying them to a large-scale KB framework means that we need to conquer the challenge of scalable learning and inference.”
Relating Things and Stuff via ObjectProperty Interactions (2014, methodology): “(6) show that this potential is an Occam razor or MDL prior, similar to [6], [13] so that the model is penalized by g l j Á jX j j when y j ¼ 1.”
Semantic structure from motion with points, regions, and objects (2012, background): “Ours w/ [17]. sky road building tree others s k y r o a d b u il d in g t r e e o t h e r s 1.0 .92 .89 .87 .56 (b) Person.”
Video Pop-up: Monocular 3D Reconstruction of Dynamic Scenes (2014, methodology): “Instead we follow the strategy of [21] and optimise the cost”
Context Driven Scene Parsing with Attention to Rare Classes (2014, background): “Context has been investigated in various semantic segmentation algorithms from co-occurrence statistics to scene categories [12, 4, 20, 7].”
Fast Approximate Energy Minimization with Label Costs (2010, background): “The application in [39] is object recognition with co-occurrance statistics.”