|
Background subtraction
is an important first step for many vision problems. It separates foreground
objects from background clutter, and enables higher-level operations, such as
tracking, object identification.
Background subtraction for natural scenes
composed of several dynamic entities is challenging. Objects of interest
often move amidst complicated backgrounds that are themselves moving, e.g.
swaying trees, other objects such as a crowd, or a flock of birds, moving
water, waves, and snow, rain, or smoke-filled environments. In such dynamic scenes,
In
this work, we introduce two different approaches to background subtraction.
Both techniques use dynamic
textures to model the video scene.
|
|
|
Project/
Results:
|
|
Generalized Stauffer-Grimson
Background Subtraction
The
model is based on a generalization of the Stauffer-Grimson background
model, where each mixture component is modeled as a dynamic texture. We
derive an on-line K-means algorithm for updating the parameters using a
set of sufficient statistics of the model.
[project]
|
|
|
|
Background Subtraction Using Discriminant
Saliency
The
algorithm is based on center-surround saliency. Background subtraction is
formulated as the complement of saliency detection, by classifying
non-salient (with respect to appearance and motion dynamics) points in
the visual field as background. The algorithm is inspired by biological
mechanisms of motion-based perceptual grouping, and extends a
discriminant formulation of center-surround
saliency proposed for static imagery. [project]
|
|
|
|
|
|
Selected Publications:
|
- Spatiotemporal Saliency in Highly Dynamic Scenes
V. Mahadevan, and N. Vasconcelos.
IEEE Trans. on Pattern Analysis and Machine Intelligence,vol. 32, no. 1, pp. 171-177, January 2010.
�IEEE,
[ps][pdf]
- On
the plausibility of the discriminant center-surround hypothesis for
visual saliency
D. Gao, V. Mahadevan, and N. Vasconcelos.
Journal of Vision, 8(7):13, 1-18, 2008. [doi:10.1167/8.7.13.]
- Background Subtraction in
Highly Dynamic Scenes
V. Mahadevan and N. Vasconcelos.
IEEE Conference on Computer Vision and Pattern Recognition (CVPR),
Anchorage, June 2008. �IEEE,
[ps][pdf]
- The discriminant center-surround hypothesis for
bottom-up saliency.
D. Gao, V. Mahadevan and N. Vasconcelos.
In Proc. Neural Information Processing Systems (NIPS),
Vancouver, Canada, 2007. [ps] [pdf]
|
![sailing1_journal_mont2_frame[1].jpg](index_files/image011.jpg)
