图像处理intro优秀培训书.pptVIP

EE569 Digital Video Processing Copyright Xin Li 2007 * Tantalizing Questions Why constrained to two frames for ME? A simplified solution: N-frame ME problem boils down to N-1 two-frame ME ones Is HVS really establishing point-to-point correspondence during motion perception? Psychological studies do not support this model New direction in joint spatio-temporal analysis What do we gain by handling N frames together? We will discuss this issue later in detail * * * * * * EE569 Digital Video Processing Copyright Xin Li 2007 EE569 Digital Video Processing Copyright Xin Li 2007 EE569 Digital Video