Effect of Saturation Arithmetic on Sum of Absolute Difference (SAD) Computation in H.264

Download Presentation

Effect of Saturation Arithmetic on Sum of Absolute Difference (SAD) Computation in H.264

Loading in 2 Seconds...

- 201 Views
- Uploaded on
- Presentation posted in: General

Effect of Saturation Arithmetic on Sum of Absolute Difference (SAD) Computation in H.264

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.

- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -

Effect of Saturation Arithmetic on Sum of Absolute Difference (SAD) Computation in H.264

Venkata Suman Sanikommu

ECE 734 Project Presentation

- Block matching between successive frames – video compression
- Find the best matching block (Motion Vector)
- Motion vector will be used to reproduce the reference frame

- Motion vectors are found by calculating minimum SAD

- Find the best matching block (Motion Vector)

- SAD Computation
- Compute (Ai – Bi) for all 16 x 16 pixels in the two blocks A and B
- Determine which Ai – Bi is less than zero and produce the absolute value in that case, else produce Ai – Bi
- Perform accumulate operation to all 16x16 absolute values.

- Overflowed values will be represented as maximum values
- Unsigned: 00…0h, FF…Fh
- Example:
- 6234h + E123h => FFFFh (saturated)

- Example:

- Unsigned: 00…0h, FF…Fh

- Use saturation arithmetic and limit the number of bits used to represent SAD values
- Reduced computation complexity
- Reduced bits for SAD
- Might affect the quality of block matching and thus motion estimation

- If min{SAD} is less than FF…Fh
- Does not affect motion estimation

- If min{SAD} is greater than FF…Fh
- Affects motion estimation
- Subset size for block matching increases
- Increased encoded file size

- We have to randomly select one for motion estimation
- Randomly selected block might not be a best match

- Subset size for block matching increases

- Affects motion estimation

- Modify H.264 SAD computation code for saturation arithmetic
- Compare the performance of H.264 video coding for modular arithmetic and Saturation arithmetic for different number of bits.
- What is the minimum number of bits required to successfully use saturation arithmetic?
- How frequently does the SAD value saturate for a given number of bits to represent?
- What is the effect of saturation on encoded file size?

Questions?