Fast software image stabilization with color registration

We present the formulation and implementation of an image stabilization system capable of stabilizing video with very large displacements between frames. A coarse-to-fine technique is applied in resolution and in model spaces. The registration algorithm uses phase correlation to obtain an initial estimate for translation between images; then Levenberg-Marquardt method for nonlinear optimization is applied to refine the solution. Registration is performed in color space, using a subset of the pixels selected by a gradient-based sub-sampling criterion. This software implementation runs at 5 Hz on non-dedicated hardware (Silicon Graphics R10000 workstation).

Authors

• Carlos Guestrin
• Fábio Gagliardi Cozman
• Eric Krotkov

Keywords

• Cameras
• Hardware
• Video sequences
• Motion estimation
• Orbital robotics
• Silicon
• Graphics
• Tracking
• Robot vision systems
• Phase estimation
• Space Model
• Color Space
• Large Displacement
• Levenberg-Marquardt Algorithm
• Registration Algorithm
• Phase Correlation
• Subset Of Pixels
• Local Minima
• Fast Fourier Transform
• Image Registration
• Error Function
• Motion Model
• Position Estimation
• Color Information
• Live Streaming
• High Frame Rate
• Feature Tracking
• Percentage Of Pixels
• Camera Motion
• Motion Compensation
• Raw Video
• Global Motion
• Rigid Frame
• Blue Sky
• Specific Hardware
• Image Registration Algorithm
• Registration Process