Scribd
Upload
23 views14 pages

Motion Estimation Techs

Uploaded by

Akilan Ramasamy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
23 views14 pages

Motion Estimation Techs

Uploaded by

Akilan Ramasamy
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 14

19L010 MULTIMEDIA

COMPRESSION TECHNIQUES
MOTION ESTIMATION TECHNIQUES

PRESENTED BY
MANEESH 21L133
MAREESHWARAN 21L134
PRIYANKA S 21L142
INTRODUCTION
• Definition: Video compression reduces the file size of a video by
removing redundancies in data. It enables efficient storage and
transmission.
• Importance: Essential for streaming, storage, and faster transmission
of videos.
• Role of Motion Estimation: A crucial technique used to compress
inter-frame redundancy by predicting motion between frames.
• The purpose of motion estimation techniques is to recover the
information by analysing the image content
BASICS OF MOTION ESTIMATION
• What is Motion Estimation?: A process of determining movement
vectors that describe the transformation from one frame to the next.
• Purpose: By estimating motion, the amount of data that needs to be
encoded is reduced.
Key Concepts:
• Frames: Reference and Target frames.
• Motion Vectors: Indicate the direction and magnitude of object
movement between frames.
Basic Idea of Motion Estimation Process Flow
BLOCK BASED MOTION
ESTIMATION (BBME)
• Concept: Divides frames into blocks and computes motion vectors for each block.
The block-based approach reduces the complexity of motion estimation by treating the video
frame as smaller, manageable units (blocks), instead of pixel-by-pixel motion tracking.
Techniques:
• Full Search Algorithm: Exhaustive comparison of every possible block match between
frames (high accuracy, high computational cost).
• Three-Step Search: A faster method that narrows down the search in three steps.
• Diamond Search: Uses diamond-shaped search patterns to locate motion vectors (fast and
effective).
• Hexagon Search: Employs hexagonal search patterns to further optimize speed.
• Advantages: Simplifies the estimation process, suitable for real-time applications.
(chatgpt mine)
BLOCK BASED MOTION
ESTIMATION (BBME)
HIERARCHICAL MOTION
ESTIMATION
 Hierarchical motion estimation is a technique used to determine motion by processing at
multiple resolution levels, typically in a pyramid structure.

Key Features
 Uses multi-scale representation (Gaussian or Laplacian pyramids).
 Coarse-to-fine approach: Starts with low resolution, refines at higher levels.
 Faster convergence due to the reduced search space at each level.

Process
Step 1: Create image pyramids (lower resolution images).
Step 2: Perform motion estimation at the coarsest level.
Step 3: Refine motion vectors at each successive higher resolution level.
PIXEL-BASED MOTION
ESTIMATION
Pixel-based motion estimation determines motion by evaluating pixel-by-
pixel changes between frames.

Key Features
Block Matching Algorithm (BMA): Divides the frame into blocks and
estimates motion for each block.
Pixel Comparison: Directly compares corresponding pixel intensities.
- Suitable for small, subtle motion detection but computationally intensive.
APPLICATION OF MOTION
ESTIMATION
Techniques
Inter-frame Compression (e.g., MPEG, H.264): Uses motion vectors to compress temporal
redundancy.
Motion Compensation: Transmits motion vectors instead of whole frames, reducing data
size.

Benefits
 Significant reduction in bit rate for video streams.
 Optimizes storage and bandwidth usage, crucial for video streaming services like Netflix,
YouTube.

Example Use Cases


 Video Streaming Platforms.
 Broadcasting and Telecommunication.
COMPARISON OF MOTION ESTIMATION
TECHNIQUES
CHALLENGES IN MOTION ESTIMATION
• Occlusion Handling: Difficult to estimate motion when objects are
partially or fully hidden.
• Complex Movements: Non-rigid or deformable motion (like flowing
water) is hard to model.
• Real-Time Processing: High computational costs make it challenging
for real-time applications like streaming.
• Energy Consumption: Optimizing for power-efficient hardware while
maintaining accuracy is crucial, especially for mobile devices.
FUTURE DIRECTIONS IN MOTION
ESTIMATION
• AI and Deep Learning: Leveraging neural networks to improve motion
prediction accuracy and reduce artifacts.
• Advanced Hardware Integration: Motion estimation algorithms
tailored for GPUs and specialized hardware.
• Real-Time Optimization: Development of more efficient algorithms
that reduce computation without compromising quality.
• Handling Non-Rigid Motion: More robust techniques to accurately
capture complex motion like human expressions or soft-body
dynamics.
CONCLUSION
Motion estimation is essential for video compression, helping
to reduce file sizes while preserving quality. While existing techniques
like block-based methods are effective, challenges such as high
computational demands remain. Future advancements in AI and
hardware optimization will further improve accuracy and efficiency.
THANK YOU !!

You might also like