Scribd
Upload
48 views

Video Coding Using Graphic Processing Unit

This document discusses using graphics processing units (GPUs) for video coding tasks. It covers several approaches that use GPUs for tasks like motion estimation in H.264 video encoding. This helps reduce computation time compared to CPUs. It also discusses using GPUs for video decoding, free viewpoint video generation from multiview video plus depth, and implementing a Wyner-Ziv video codec for faster decoding. Throughout, the goal is to improve coding efficiency and reduce computational complexity of video coding standards.

Uploaded by

Naresh Devineni
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
48 views

Video Coding Using Graphic Processing Unit

This document discusses using graphics processing units (GPUs) for video coding tasks. It covers several approaches that use GPUs for tasks like motion estimation in H.264 video encoding. This helps reduce computation time compared to CPUs. It also discusses using GPUs for video decoding, free viewpoint video generation from multiview video plus depth, and implementing a Wyner-Ziv video codec for faster decoding. Throughout, the goal is to improve coding efficiency and reduce computational complexity of video coding standards.

Uploaded by

Naresh Devineni
Copyright
© Attribution Non-Commercial (BY-NC)
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 18

VIDEO CODING USING GRAPHIC

PROCESSING UNIT

BY:
Naga Mounika Reddy - 2538148
Neelima Paleti - 2538429
Jhansirani Nuguri - 2535615
Sundar Sairam. Yadlapalli - 2535614
Graphic Processing Unit
• What is GPU?

• Goal of GPU.

• Applications of GPU.
Video Coding
• What is video coding?

• Goal of video coding.

• Applications of video coding.

• Standards of video coding.


H.264 Video Coding
(MPEG-4 Advanced Video Coding)

MPEG.2 MPEG.4 H.264


DEPTH MAP ENHANCED MACROBLOCK
PARTITIONING FOR H.264 VIDEO CODING OF
COMPUTER GRAPHICS CONTENT

• Processing of 3D graphics is computationally


expensive.
• They pose problems like time delay and over
loading of encoder.
SKY BOX

 Render content information is used to detect sky box


regions.

 Suitable macroblock partitioning is predicted.

 Projection matrices are used to calculate the


corresponding motion vector directly.
REAL-TIME FREE-VIEWPOINT VIEWER FROM MULTIVIEW VIDEO
PLUS DEPTH
REPRESENTATION CODED BY H.264/AVC MVC EXTENSION

o Increases view point flexibility.

o Free view point navigation.

o Real time decoding of multiple videos.

o It can generate free view point images at 30fps.


Block Based Parallel Motion Estimation Using Programmable Graphics
Hardware

• Overcomes dependency problem.


• Used for full search and fast search.
• Forty five faster than our SDVTD optimized
CPU implementation.
High-Level
diagram of
modern graphics
pipeline

High-Level block
diagram of GPU-
based
motion estimation
Fast Motion Estimation on Graphics
Hardware for H.264 Video Encoding

 Higher coding efficiency.


Increase the encoding complexity ,which is a time consuming task for
CPU.

 GPU based approach for motion estimation that adapts a small diamond
search.

 Significant reduction of computation time and a competitive encoding


quality compared to a CPU UMHexagonS implementation while enabling
the CPU to process other encoding tasks in parallel.
H.264 Video Decoding Compatible with
Vector Graphics
 Decode video H.264 stream.

 Use advanced vector graphics pipeline instead of native


decoder.

 Accelerate video at resolutions between QCIF to full HD


without loss or anu drift of quality
Video Mapping at Functional level
Fast Implementation of Wyner-Ziv Video Codec using
GPGPU

• Implement low complexity video encoder.

• Provides a problem of decoding complexity.

• Focus on fast implementation of the Slepian-Wolf decoder using the


CUDA (Compute Unified Device Architecture).

• Slepian-Wolf decoding show that our scheme is 4~5 times faster in


QCIF size sequences and 15~20 times faster in CIF size sequences.
Conclusion
• Two standards of video coding
 H.264
 Wyner Ziv
• Each standard rectifies some
problems in video coding like
 Decreasing complexity and
computational time.
 Avoids dependency.
References
 Fast Motion Estimation on Graphics Hardware for H.264 Video Encoding, Martin Schwalb,
Ralph Ewerth, and Bernd Freisleben, Member, IEEE.
 Block Based Parallel Motion Estimation Using Programmable Graphics Hardware, M.C. Kung,
Oscar C. Au, P.H.W. Wong, Chun Hung Liu, Department of Electronic and Computer
Engineering, The Hong Kong University of Science and Technology.
 Real-time free-viewpoint viewer from multiview video plus depth representation coded by
H.264/AVC MVC extension, Shinya Shimizu, Hideaki Kimata, and Yoshimitsu Ohtani, NTT
Cyber Space Laboratories, NTT Corporation
 Depth map enhanced macroblock partitioning for H.264 video coding of computer graphics
content, Philipp Fechteler and Peter Eisert, Fraunhofer Institute for Telecommunications -
Heinrich-Hertz-Institute, Image Processing Department.
 H.264 Video Decoding Compatible with Vector Graphics, Danilo Pau, Francesco Mariotti,
Marco Orlandini , Luca Pezzoni, Advanced System Technology, STMicroelectronics, Agrate,
Italy.
 Fast Implementation of Wyner-Ziv Video Codec using GPGPU, Ryanggeun Oh, Jongbin Park,
and Byeungwoo Jeon, Department of Mobile Systems Engineering, School of Information and
Communication Engineering, Sungkyunkwan University, Korea.

You might also like