homework-1

Motion Estimation

Block Matching Algorithm

Follow the report »

Usage

BlockMatching() class is written in Python. It takes 5 arguments:

• dfd : {0:MAD, 1:MSE} Displaced frame difference
• blockSize : (sizeH,sizeW)
• searchMethod : {0:Exhaustive, 1:Three-Step}
• searchRange : (int) +/- pixelwise range
• motionIntensity: True (default)
  Normalization for motion vector intensities. Assigns 255 to the largest amplitude motion vector. Also, there is a threshold that intensity value cannot be less than 100.

In main script, there are two more parameters to control the predictions:

• predict_from_prev : False (default)
• N: 5 (default)
  If the predictions made from previous predicted frames, anchor is updated after N frames.

Initialize an object of BlockMatching() class with an arbitrary name, bm() is suggested.

bm = BlockMatching(dfd=dfd,
          blockSize=blockSize,
          searchMethod=searchMethod,
          searchRange=searchRange,
          motionIntensity=False)

Then use step() method to run the program. After execution, you can reach the generated motion field via bm.motionField and the predicted anchor bm.anchorP properties.

bm.step(anchor,target)
motionField = bm.motionField
anchorP = bm.anchorP

Use visualize() method of Video() class to put the 4 frames together to show a collage. Here, a and t arguments are the frame numbers of the anchor and target frames, respectively.

collage = video.visualize(anchor,target,motionField,anchorP,text,a,t)

Demo