280 likes | 617 Views
Spatiotemporal Saliency Detection and Its Applications in Static and Dynamic Scenes. IEEE TCSVT 2011 Wonjun Kim Chanho Jung Changick Kim. Outline. Introduction Proposed Method Experiment Result Application Conclusion. Introduction. Problem occurs when background is highly textured.
E N D
Spatiotemporal Saliency Detection and Its Applications in Static and Dynamic Scenes IEEE TCSVT 2011 Wonjun Kim Chanho Jung Changick Kim
Outline • Introduction • Proposed Method • Experiment Result • Application • Conclusion
Introduction • Problem occurs when background is highly textured
Proposed Method • Feature Representation • Edge orientation histogram (EOH) • Color orientation histogram (COH) • Temporal Feature • Self-ordinal Measure • Saliency Map • Scale-invariant Saliency Map
Edge Orientation Histogram (EOH) • Compute the edge orientation of every pixel in the local region center at the pixel • Quantized into K angle in the range of [,] • Compute the histogram of edge orientation local region m(x,y,n):edge magnitude (x,y,n):quantized orientation
Color Orientation Histogram (COH) • Quantize the angle in HSV color space in the range of [,] into H angles • Compute the histogram of color orientation s(x,y,n):saturation value (x,y,n):quantized hue value
Temporal Feature • Compute the intensity differences between frames • Feature at the pixel of frame P :total number of pixels in local region j :index of those pixels in P :user-defined latency
Self-ordinal Measure • Define a 1(K+1) rank matrix by ordering the elements of EOH(COH) ex:
Saliency Map of Edge and Color • Compute the distance from the rank matrix of center region to surrounding regions Saliency Map of Edge Saliency Map of Color N :total number of local regions in a center-surround window ,:maximum distance between two rank matrices
Spatial Saliency Map • Combine the edge and color saliency
Combining with Temporal Saliency • Compute the SAD of temporal gradients between center and the surrounding regions • Combine the spatial and temporal saliency
Scale-invariant Saliency Map • Combine 3 different scales of saliency Map(3232, 6464, 128128) 3232 128128 6464
Experiment Result • Static Images • Video Sequences
Experiment Result • Static Image • Local region = 55 • center-surround window = 77 • K = 8, H= 6 • = 40, = 24 • Video Sequence • = 49 • Speed: 23ms per frame (43 fps)
Application • Image Retargeting • Moving Object Extraction
Moving Object Detection • G:the set of salient pixels in the ground truthimage • P:salient pixels in the binarized object map • Card(A):the size of the set A
Conclusion • Ordinal signature can tolerate more local feature distribution than sample values. • The proposed scheme performs in real-time and can be extended in both static and dynamic scenes.