This paper presents a spatial-semantic modeling system featuring automated learning of object-place relations from an online annotated database, and the application of these relat...
Pooja Viswanathan, David Meger, Tristram Southey, ...
This paper presents a novel approach to object recognition involving a sparse 2D model and matching using video. The model is generated on the basis of geometry and image measurab...
Humera Noor, Shahid H. Mirza, Yaser Sheikh, Amit J...
Automatic image orientation detection for natural images is a useful, yet challenging research area. Humans use scene context and semantic object recognition to identify the corre...
Most digital still color cameras use a single electronic sensor (CCD or CMOS) overlaid with a color filter array. At each pixel location only one color sample is taken, and the ot...
A reliable system for visual learning and recognition should enable a selective treatment of individual parts of input data and should successfully deal with noise and occlusions....