A Bayesian hierarchical model for learning natural scene categories

By Li Fei-Fei, Pietro Perona. 2005

4,187 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Probabilistic & Bayesian Methods

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 Probabilistic & Bayesian Methods

Citations in the corpus

LabelMe: A Database and Web-Based Tool for Image Annotation (2008, background): “More recently, it has become popular to apply “bag of word” techniques to discover object categories from images which are known to contain the object [10, 5, 12, 11, 9].”
Seeing Out of tHe bOx: End-to-End Pre-training for Vision-Language Representation Learning (2021, background): “VideoBERT [36] and the bag of words [13] literature also use vector quantization to represent visual information.”
Recognizing indoor scenes (2009, methodology): “Most images within each category average to a unifor m field due to the large variability within each scene category (this is in contrast with Caltech 101 or the 15 scene categories dat aset [9, 3, 7].”
Improving Interpretability of Deep Neural Networks with Semantic Information (2017, methodology): “As proven in previous work [13, 7], topic models can extract semantically meaningful concepts (or themes) that are useful for visual analysis tasks.”
On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation (2015, background): “image search [1], robotics [2], medical imaging [3], object detection in radar images [4] or face detection [5].”
This looks like that: deep learning for interpretable image recognition (2018, background): “It relates to traditional “bag-of-visual-words” models used in image recognition [21, 6, 17, 38, 31].”
Nonparametric Divergence Estimation with Applications to Machine Learning on Distributions (2011, methodology): “Features are extracted as in Fei-Fei and Perona (2005). Points are firstly sampled on a uniform grid with interval 5.”
SUN Database: Exploring a Large Collection of Scene Categories (2014, background): “For comparison with previous works, we show results on the 15 scene categories dataset (Oliva and Torralba 2001; Lazebnik et al. 2006; Fei-Fei and Perona 2005) in Fig.”
Using Multiple Segmentations to Discover Objects and their Extent in Image Collections (2006, background): “Recently, some success has also been reported in discovering object and scene categories [7, 17, 21] by borrowing tools from the statistical text analysis community.”
Context Encoding for Semantic Segmentation (2018, methodology): “Then a visual vocabulary (dictionary) is often learned and the global feature statistics are described by classic encoders such as Bag-of-Words (BoW) [8, 13, 26, 46], VLAD [25] or Fisher Vector [44].”
Discovering objects and their location in images (2005, background): “These results demonstrate that we can successfully build object class models from an unsupervised analysis of images.”
SIFT Flow: Dense Correspondence across Scenes and Its Applications (2011, methodology): “We also apply SIFT flow back to the regime of traditional image alignment, such as satellite image registration and face recognition.”
Data-Driven Grasp Synthesis—A Survey (2013, background): “, a common texture [81] or shape [82, 83], the occurrence of specific local features [84, 85] or their specific spatial constellation [86, 87].”
An empirical study of context in object detection (2009, background): “Semantic context might indicate the kind of event, activity, or other scene category is being depicted [25, 8] or the presence (and possibly location) of objects in the scene [28, 1].”