PCV @ OSU

Photogrammetric Computer Vision Laboratory

Social computing

Social computing

Learning and analysis of the social structure in a scene from sensory information.

Tracking

Tracking

We have extensively worked on the object tracking problem and wrote the only comprehensive survey in the literature. Our methods include a wide range of techniques, such as point tracking, contour tracking and kernel tracking.

Clustering

Clustering

Clustering in images, video and other multi physics and multi-dimensional data using various computational models.

Action recognition

Action recognition

Modeling and recognition of human actions. Our primary emphasis has been on the invariance to viewpoints and performing actors.

Projective Geometry

Projective Geometry

3D recovery and camera calibration methods using different techniques including but not limieted to plane projective geometry from uncalibrated cameras, lines and motion.

Shape representations

Shape representations

Representation of planar shapes using feautures invariant to viewpoint, such that are invariant to affine and projective transformation.

Navigation

Navigation

Personal navigation by fusion of sensor suit which includes stero camera, IMU, pressure sensors and radio beacons.

Human Faces

Human Faces

We have worked on a range of other problems that include the facial motion synthesis, pose recovery, shape and albedo from shading and face recognition.

Tag Cloud

Action Recognition; Social Network; Active Contour; Object Recognition; Shape Recovery; Segmentation; Tracking; View Geometry; Navigation;

PCVLab @ Twitter

Follow us on twitter

Director

Alper Yilmaz, PhD

Asst. Prof. Civil, Env. & Geodetic Eng.
Asst. Prof. Computer Sci. & Eng. (courtesy)
Curriculum Vitae

Office: 214C Bolz Hall
Email: yilmaz.15@osu.edu
Phn: 614-247-4323
H-Index (research impact)

Recent Articles

H. Lee and A. Yilmaz, Intrinsic Color from a Single Outdoor Image, ASPRS Annual Conference. Sacramento, CA, 2012

L. Ding and A. Yilmaz, Inferring Social Relations from Visual Concepts, IEEE International Conf. on Computer Vision (ICCV). Barcelona, Spain, 2011

P. Srestasathiern and A. Yilmaz, Planar Shape Representation and Matching Under Projective Transformation, Computer Vision and Image Understanding, Vol. 115. No. 11, pp. 1525-1535, 2011

J. Jiang and A. Yilmaz, Good Features to Track: A View Geometric Approach, IEEE Workshop on Mobile Vision (in conjunction with ICCV). Barcelona, Spain, 2011

L. Ding, A. Yilmaz and R. Yan, Interactive Image Segmentation Using Dirichlet Process Multiple View Learning, IEEE Trans. on Image Processing (accepted for publication), 2011