Spatiograms versus histograms for region-based tracking

By Stan Birchfield, Sriram Rangarajan. Computer Vision and Pattern Recognition, 2005

497 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/9b2b3e7460f86d252fa5e0bff6b83fc235fbd73e

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

Citations in the corpus

Object Tracking Benchmark (2015, methodology): “Instead of relying on pixel-wise statistics, Birchfield and Rangarajan [11] proposed the spatiogram to capture both the statistical properties of pixels and their spatial relationships.”
A survey of appearance models in visual object tracking (2013, unknown): “The goal of joint spatial-color modeling is to describe the distribution properties of object appearance in a joint spatial-color space (e.g., (x, y, R, G, B) [Yang et al. 2005; Georgescu and Meer 2004; Birchfield and Rangarajan 2005]).”
Evaluating Appearance Models for Recognition, Reacquisition, and Tracking (2007, background): “Also noteworthy are multi-resolution histograms [16], image signatures [24], and spatiograms [3].”
Human activities recognition using depth images (2013, background): “More importantly, recent depth cameras (e.g. Kinect) offer substantial depth resolutions of a few centimeters and thus provide very good estimation of the three-dimensional geometry of the scene.”
Multi-target tracking by on-line learned discriminative appearance models (2010, unknown): “, Haarlike features [24], SIFT-like features [12, 19], orientation histograms [10, 17], spatiograms [6], are integrated in this framework.”
Modeling spatial layout with fisher vectors for image categorization (2011, methodology): “Our work is also related to [1] which employed first and second order spatial moments associated with bins of color histograms to derive an improved representation for mean-shift tracking.”
Distribution fields for tracking (2012, methodology): “Other additions are higher order statistics [4] or temporal information [5], and using feature selection [7, 16, 26].”