Course Outline for Digital Video Processing (EE392J)

Course Outline

• Introduction/Course Overview
• Analog video and NTSC television
• Spatio-temporal sampling; Sampling structure conversion (without using motion) 
• Motion Analysis
      Real versus apparent motion
      Spatial-temporal constraint methods (optical flow equation)
      Block-matching methods
      Mesh-based methods
      Region-based (parametric) motion modeling
      Motion segmentation and layered video representations
• Video Processing
      Motion-compensated (MC) filtering
      Noise reduction
      Signal recovery and general inverse problems
          Restoration (deblurring)
          Superresolution, Mosaicing
      Deinterlacing
      Frame-rate conversion (MC-Interpolation)
• Video Watermarking
• Video Compression
      Frame-based compression (principles behind MPEG-1/2)
      Scalable or layered frame-based compression
      Object-based compression (principles behind MPEG-4) 
• Video communication
      Video streaming and error-resilience
      Efficient processing of compressed video
      Digital TV