Fast Discriminative Visual Codebooks using Randomized Clustering Forests

By Frank Moosmann, Bill Triggs, Frédéric Jurie. Neural Information Processing Systems, 2006

556 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/c2d3746a1f755928b5011932285d686eb5a9127b

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

Citations in the corpus

Image Classification using Random Forests and Ferns (2007, background): “plied to object recognition in [20, 24] but only for a relatively small number of classes.”
Gaussian Processes for Object Categorization (2010, methodology): “Early models captured appearance and shape variation in a generative probabilistic framework [11], but more recent techniques have typically exploited methods based on SVMs or Nearest Neighbors in a bag-of-visual-words feature space [36,30,49,9].”
Modeling spatial layout with fisher vectors for image categorization (2011, background): “Random forests, because of their hierarchical structure, are also good candidates for handling large visual vocabularies [4, 13].”
Empowering Visual Categorization With the GPU (2011, methodology): “Faster to compute descriptors (such as SURF [13], [14]) and indexing mechanisms (tree-based codebooks [15], [16]) have been developed.”
On-line Random Forests (2009, background): “RFs have demonstrated to be better or at least comparable to other stateof-the-art methods in both classification [6] [4], semantic segmentation [21], real-time keypoint recognition [13] and clustering applications [14].”
Deep Neural Networks and Tabular Data: A Survey (2021, methodology): “This embedding based on decision tree leaves was ﬁrst proposed in [101] for the random forest algorithm.”
Learning to Search Efficiently in High Dimensions (2011, methodology): “The data structured considered in this paper is motivated by the success of vocabulary tree method in image retrieval [18, 27, 15], which has been adopted in modern image search engines to find near duplicate images.”
Class-specific Hough forests for object detection (2009, unknown): “Does offset supervision matter? Quite a few previous approaches have used random forests as discriminative codebooks [12, 13, 14, 20, 25].”
Semantic texton forests for image categorization and segmentation (2008, methodology): “Semantic texton forests, the main contribution of this paper, extend [20] in three ways: (i) we learn directly from image pixels, bypassing the expensive step of computing image descriptors; (ii) while [20] use the learned decision forest only for clustering, we also use it as a classifier, which en”
Understanding Random Forests: From Theory to Practice (2014, methodology): “80 random forests From a more practical point of view, such forest-based transforms φ and φ′ have proven to be helpful and efficient embeddings, e.g., when combined with linear methods or support vector machines [Moosmann et al., 2006; Marée et al., 2013].”
Combining randomization and discrimination for fine-grained image categorization (2011, methodology): “Random forests have been used successfully in many vision tasks such as object detection [1], segmentation [17] and codebook learning [15].”
Efficient Human Pose Estimation from Single Depth Images (2013, background): “Randomized decision trees and forests [46], [47], [48], [49], [50] have proven fast and effective multiclass classifiers for many tasks [44], [51], [52], [50], and can be implemented efficiently on the GPU [12].”
Real-Time Online Action Detection Forests Using Spatio-Temporal Contexts (2016, background): “This coincides the split criterion of regression RFs [41, 9, 28] (Fig.”
Spacetime Forests with Complementary Features for Dynamic Scene Recognition (2013, unknown): “, [4, 19, 23, 36].”