SlideShare a Scribd company logo
1 of 28
Dynamic Adaptive Point Cloud Streaming
ACM Packet Video 2018 (PV’18)
Mohammad Hosseini
University of Illinois at Urbana-Champaign (UIUC)
shossen2@Illinois.edu
Christian Timmerer
Alpen-Adria-Universitat Klagenfurt, Bitmovin Inc.
christian.timmerer@itec.uni-klu.ac.at
2 /28
Introduction
• 3D point clouds are set of points represented in the 3D space
• Recently emerged as an advanced representation of immersive media
• Can be captured using camera arrays and depth sensors in various setups
3 /28
• Point clouds are very bulky!
– Can consist of billions of points
– Highly resource intensive
– The data transmission rate in raw point-cloud frames in an application with 30 fps
can be as high as 6 Gbps
Motivation
Goal:
How to achieve a balance between the streaming quality and the available resources
on the display devices without much negative on user’s viewing experience
4 /28
Related Work
• Dynamic Adaptive Streaming over HTTP (MPEG-DASH)
– A client chooses a video segment with the appropriate quality (bit rate, resolution, etc.) based on its
constrained resources such as bandwidth
– The video content is stored on an HTTP server, and is accompanied by a Media Presentation
Description (MPD)
– We extend the semantics of MPEG-DASH towards a dynamic point cloud streaming system
• To enable view-aware and bandwidth-aware adaptation through transmitting different point cloud
representations relative to the user’s resources
• Point Cloud Compression (PCC)
– Existing works are mostly centered around point cloud compression using geometry-based approaches
– On the MPEG side, a new work item has been initiated for Point Cloud Compression, or in short, MPEG
PCC to cover topics in lossy and lossless point cloud compression
5 /28
• DASH-PC: Extend DASH towards adaptive point cloud streaming
• Point-cloud density reduction based on distance-performance-quality trade-offs
Proposed Approach
15K
points
30K
points
150K
points
1M
points
Storage, Processing, Power, etc.
6 /28
• Quality adaptation based on distance
• Human Visual Acuity!
– 60 Pixel per degree for adults
Proposed Approach
D … distance of a user from the display screen
S … scale factor of a point cloud object
D’ … distance of the camera position from the centroid of the object’s bounding box
7 /28
Proposed Approach
Low Density
(Rep. 3)
Medium Density
(Rep. 2)
High Density
(Rep. 1)
8 /28
DASH-PC: Architecture Overview
9 /28
• Similar to DASH, the adaptation feature of DASH-PC is described
by an XML-formatted manifest containing metadata
• Hierarchical manifest, divided into separate PC frames. Each frame
including a variety of adaptation sets:
– Multiple quality alternatives
– Index of each frame
– The frame’s HTTP location
– Level of Density (LoD) representations
DASH-PC: Manifest
10 /28
• An example manifest
DASH-PC: Manifest
11 /28
• Re-use some of the DASH mime types for a better compliance with MPEG-DASH
– @format: specifies the point cloud container format (e.g., ply)
– @encoding: specifies the point cloud encoding (ASCII or binary)
– @frames: specifies the number of point cloud frames followed
– @type: shows if the manifest may be updated during a session (’static’ (default) or
’dynamic’)
– BaseURL: specifies the base HTTP URL of the point cloud
DASH-PC: Manifest
12 /28
• Frame
– Typically represents a single point cloud model
– For each frame, a consistent set of multiple adaptation sets of that point cloud model
are available
– @id: a unique identifier of the frame
– BaseURL: specifies the base HTTP URL of the frame
– AdaptationSet: specifies the sub-clause adaptation sets
DASH-PC: Manifest
13 /28
• AdaptationSet:
– Consists of multiple interchangeable quality versions of a point cloud frame
– It can carry additional attributes such as:
• Density sub-sampling methods
• Data structure used
• Compression techniques
• Viewport information
– @id: a unique identifier of the adaptation set
– BaseURL: specifies the base HTTP URL of the adaptation set
– Representation: specifies the sub-clause representations
DASH-PC: Manifest
14 /28
• Representation
– A single point cloud representation is sufficient for rendering and visualization of a model
– A client can switch from one representation to another at any time
• To adapt to its varying resources and parameters such as network bandwidth, energy budget, visual
preferences, etc.
– @id: unique identifier of the representation
– BaseURL: specifies the base HTTP URL of the representation
– @density: specifies the density of the point cloud model in terms of the number of
points
– @size: specifies the size of the point cloud model in terms of the storage volume
– Segment: specifies the sub-clause segments
DASH-PC: Manifest
15 /28
• Segment
– Describe sub-models based on spatial segmentation which can be used to retrieve
partial point cloud models
– For better compliance with the features of viewport adaptations
– @id: unique identifier of the adaptation set
– BaseURL: specifies the HTTP URL of the segment
– @density: specifies the density of the point cloud model in terms of the number of
points
– @size: specifies the size of the point cloud model in terms of the storage volume
DASH-PC: Manifest
16 /28
• Proposed three different point-cloud density reduction algorithms
• Based on the concept of spatial clustering of neighbor points in the 3D space and
sampling points in each cluster
Density Reduction
17 /28
• Algorithm 1:
– Clusters are simply constructed through points geometrically being sorted in the 3D space
– Each cluster with near-minimum spatial Euclidean distance of consisting points
• Algorithm 2:
– Use the notion of histogram, and construct a data structure called “low-resolution point cloud density tree”
• Algorithm 3:
– Enhanced version of Algorithm 2
Density Reduction: Algorithms Overview
18 /28
• Developed tools and simple APIs for users to construct representations based
on sub-sampling and scaling
• subsample(int percentage, String inputpc, bool isIterative)
– Generates a sub-sampled point cloud model with the percentage integer specifying the density of the
output point cloud model relative to the input point cloud
– isIterative flag specifies if the process is iterated over multiple successive frames
• scale(int percentage, String inputpc, bool isIterative)
– Generates a scaled point cloud model given the scaling ratio
• optimize(int distance, int scale, int inputpc, bool isIterative)
– Automatically generates a sub-sampled point cloud with optimum PPI density given the distance (in inch)
and/or the scale
Density Reduction: User API
19 /28
• Used Point Cloud Library (PCL 1.8.1) to develop a dynamic point cloud visualization and
streaming prototype
• Used four different sequences provided by JPEG Pleno Database containing 8i voxelized
dynamic point cloud sequences
– longdress, loot, redandblack, and soldier
– In each sequence, the full object is captured by 42 RGB cameras at 30 fps, over a 10 second period
Experimental Setup
20 /28
• Intel Xeon E5520 64-bit machine with quad-core 2.27 GHz CPU, 12 GB RAM, and
Gallium 0.4 on AMD GPU
• Ubuntu 16.04 LTS
• Server-client setup
– Used our machine as the local HTTP streaming server to filter the negative impacts of network
latency and bandwidth variations on the experimental results
– To focus purely on the amount of bandwidth savings, rendering performance, and objective quality
measurement
Experimental Setup
21 /28
• Applied four sub-sampling ratios to the test sequences, and prepared 5 different
representations in total for each point cloud frame
– REP1 to REP5 representing highest to lowest density, given sub-sampling ratios of 1 to 5
• Measured three different factors:
– Total bandwidth usage  directly correlated to the sub-sampling ratio
• Total bandwidth saving of approximately 50% and 80% was achieved in raw point cloud frames given sub-sampling ratios
of 2 and 5, respectively
– Rendering performance (FPS)
– Objective quality (PSNR)
• Compared our results with the baseline scenario where no adaptation is employed
• Each trial of our experiment was run for a total of 30 consecutive test sequences
• Repeated each experiment 6 times
– To ensure that the standard deviations for variable metrics are within acceptable limits
Evaluation
22 /28
• Measured the client’s rendering performance in terms of the average FPS
– Measured FPS continuously for 100 times during each session
• Recorded the data when adaptations are applied
– Lower representations (REP2 and REP5 with sub-sample ratios of 2 and 5) compared
to the baseline case where no adaptation is applied (REP1)
Evaluation: Rendering Performance
23 /28
• Rendering Performance (FPS)
– Results for only a subset of our experiments on all of our benchmarks under Algorithm 1
– Our adaptations can significantly improve the FPS up to more than 10x
Evaluation: Rendering Performance
24 /28
• Objective Quality (PSNR)
– Extensive measurements on the objective quality using PSNR of point-to-point
distortions to verify our assumptions for visual acuity
– Used the official MPEG PCC quality metric software
• Compares an original point cloud with an adapted model and provides
numerical values for point cloud PSNR
– Averaged the PSNR measurements over all dynamic point cloud frames, and
collected the experimental results
Evaluation: Objective Quality
25 /28
• Objective Quality (PSNR)
– PSNR under 2x, 4x, and 8x scale-down ratios
– Confirms our earlier assumptions in regards to the effects of scaling point clouds on the human’s
visual acuity
Evaluation: Objective Quality
26 /28
• Visual View
– Illustrating three consecutive frames from highest representation (REP1) (left) to the lowest
representation (REP5) (right)
Evaluation: Visual View
Visual comparison of a specific frame within Red&Black under 3 sample representations:
(left) REP1, (middle) REP2, (right) REP5.
27 /28
Conclusion and Future Work
• Conclusions:
– Proposed DASH-PC, a dynamic point cloud streaming adaptation system to tackle the high bandwidth
and processing requirements of dense point clouds
– Studied the semantic link of adaptive streaming with a user’s visual acuity, limited bandwidth, and
performance overhead
– Initial experimental results show that our adaptations can significantly save bandwidth and improve
rendering performance on point cloud streaming with negligible noticeable quality impacts
• Future Work:
– Study various rate adaptation algorithms and evaluate their quality impacts
28 /28
Questions

More Related Content

What's hot

MULTIMEDIA COMMUNICATION & NETWORKS
MULTIMEDIA COMMUNICATION & NETWORKSMULTIMEDIA COMMUNICATION & NETWORKS
MULTIMEDIA COMMUNICATION & NETWORKSKathirvel Ayyaswamy
 
AI On the Edge: Model Compression
AI On the Edge: Model CompressionAI On the Edge: Model Compression
AI On the Edge: Model CompressionApache MXNet
 
Versatile Video Coding – Video Compression beyond HEVC: Coding Tools for SDR ...
Versatile Video Coding – Video Compression beyond HEVC: Coding Tools for SDR ...Versatile Video Coding – Video Compression beyond HEVC: Coding Tools for SDR ...
Versatile Video Coding – Video Compression beyond HEVC: Coding Tools for SDR ...Förderverein Technische Fakultät
 
Adaptive Media Streaming: The Role of Standards
Adaptive Media Streaming: The Role of StandardsAdaptive Media Streaming: The Role of Standards
Adaptive Media Streaming: The Role of StandardsAlpen-Adria-Universität
 
Message and Stream Oriented Communication
Message and Stream Oriented CommunicationMessage and Stream Oriented Communication
Message and Stream Oriented CommunicationDilum Bandara
 
Shared Memory Centric Computing with CXL & OMI
Shared Memory Centric Computing with CXL & OMIShared Memory Centric Computing with CXL & OMI
Shared Memory Centric Computing with CXL & OMIAllan Cantle
 
HEVC Definitions and high-level syntax
HEVC Definitions and high-level syntaxHEVC Definitions and high-level syntax
HEVC Definitions and high-level syntaxYoss Cohen
 
Dichotomy of parallel computing platforms
Dichotomy of parallel computing platformsDichotomy of parallel computing platforms
Dichotomy of parallel computing platformsSyed Zaid Irshad
 

What's hot (14)

Multimedia chapter 5
Multimedia chapter 5Multimedia chapter 5
Multimedia chapter 5
 
MULTIMEDIA COMMUNICATION & NETWORKS
MULTIMEDIA COMMUNICATION & NETWORKSMULTIMEDIA COMMUNICATION & NETWORKS
MULTIMEDIA COMMUNICATION & NETWORKS
 
AI On the Edge: Model Compression
AI On the Edge: Model CompressionAI On the Edge: Model Compression
AI On the Edge: Model Compression
 
Versatile Video Coding – Video Compression beyond HEVC: Coding Tools for SDR ...
Versatile Video Coding – Video Compression beyond HEVC: Coding Tools for SDR ...Versatile Video Coding – Video Compression beyond HEVC: Coding Tools for SDR ...
Versatile Video Coding – Video Compression beyond HEVC: Coding Tools for SDR ...
 
Adaptive Media Streaming: The Role of Standards
Adaptive Media Streaming: The Role of StandardsAdaptive Media Streaming: The Role of Standards
Adaptive Media Streaming: The Role of Standards
 
Message and Stream Oriented Communication
Message and Stream Oriented CommunicationMessage and Stream Oriented Communication
Message and Stream Oriented Communication
 
Green IT
Green ITGreen IT
Green IT
 
H261
H261H261
H261
 
Multimedia communication networks
Multimedia communication networksMultimedia communication networks
Multimedia communication networks
 
Shared Memory Centric Computing with CXL & OMI
Shared Memory Centric Computing with CXL & OMIShared Memory Centric Computing with CXL & OMI
Shared Memory Centric Computing with CXL & OMI
 
MHEG
MHEGMHEG
MHEG
 
HEVC Definitions and high-level syntax
HEVC Definitions and high-level syntaxHEVC Definitions and high-level syntax
HEVC Definitions and high-level syntax
 
H.263 Video Codec
H.263 Video CodecH.263 Video Codec
H.263 Video Codec
 
Dichotomy of parallel computing platforms
Dichotomy of parallel computing platformsDichotomy of parallel computing platforms
Dichotomy of parallel computing platforms
 

Similar to Dynamic Adaptive Point Cloud Streaming

A methodology for full system power modeling in heterogeneous data centers
A methodology for full system power modeling in  heterogeneous data centersA methodology for full system power modeling in  heterogeneous data centers
A methodology for full system power modeling in heterogeneous data centersRaimon Bosch
 
RECAP: The Simulation Approach
RECAP: The Simulation ApproachRECAP: The Simulation Approach
RECAP: The Simulation ApproachRECAP Project
 
[EWiLi2016] Enabling power-awareness for the Xen Hypervisor
[EWiLi2016] Enabling power-awareness for the Xen Hypervisor[EWiLi2016] Enabling power-awareness for the Xen Hypervisor
[EWiLi2016] Enabling power-awareness for the Xen HypervisorMatteo Ferroni
 
GPU Renderfarm with Integrated Asset Management & Production System (AMPS)
GPU Renderfarm with Integrated Asset Management & Production System (AMPS)GPU Renderfarm with Integrated Asset Management & Production System (AMPS)
GPU Renderfarm with Integrated Asset Management & Production System (AMPS)Budianto Tandianus
 
Multi Processor Architecture for image processing
Multi Processor Architecture for image processingMulti Processor Architecture for image processing
Multi Processor Architecture for image processingideas2ignite
 
Application Profiling at the HPCAC High Performance Center
Application Profiling at the HPCAC High Performance CenterApplication Profiling at the HPCAC High Performance Center
Application Profiling at the HPCAC High Performance Centerinside-BigData.com
 
APSys Presentation Final copy2
APSys Presentation Final copy2APSys Presentation Final copy2
APSys Presentation Final copy2Junli Gu
 
RAMSES: Robust Analytic Models for Science at Extreme Scales
RAMSES: Robust Analytic Models for Science at Extreme ScalesRAMSES: Robust Analytic Models for Science at Extreme Scales
RAMSES: Robust Analytic Models for Science at Extreme ScalesIan Foster
 
CAD theory presentation.pptx .
CAD theory presentation.pptx                .CAD theory presentation.pptx                .
CAD theory presentation.pptx .Athar739197
 
Three Challenges in Reliable Data Transport over Heterogeneous ...
Three Challenges in Reliable Data Transport over Heterogeneous ...Three Challenges in Reliable Data Transport over Heterogeneous ...
Three Challenges in Reliable Data Transport over Heterogeneous ...Videoguy
 
Software-Defined Systems for Network-Aware Service Composition and Workflow P...
Software-Defined Systems for Network-Aware Service Composition and Workflow P...Software-Defined Systems for Network-Aware Service Composition and Workflow P...
Software-Defined Systems for Network-Aware Service Composition and Workflow P...Pradeeban Kathiravelu, Ph.D.
 
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degree
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degreeThe UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degree
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degreePradeeban Kathiravelu, Ph.D.
 
Standardising the compressed representation of neural networks
Standardising the compressed representation of neural networksStandardising the compressed representation of neural networks
Standardising the compressed representation of neural networksFörderverein Technische Fakultät
 
Mirabilis_Design AMD Versal System-Level IP Library
Mirabilis_Design AMD Versal System-Level IP LibraryMirabilis_Design AMD Versal System-Level IP Library
Mirabilis_Design AMD Versal System-Level IP LibraryDeepak Shankar
 
byteLAKE's expertise across NVIDIA architectures and configurations
byteLAKE's expertise across NVIDIA architectures and configurationsbyteLAKE's expertise across NVIDIA architectures and configurations
byteLAKE's expertise across NVIDIA architectures and configurationsbyteLAKE
 
XLcloud 3-d remote rendering
XLcloud 3-d remote renderingXLcloud 3-d remote rendering
XLcloud 3-d remote renderingMarius Preda PhD
 
Concurrent Replication of Parallel and Distributed Simulations
Concurrent Replication of Parallel and Distributed SimulationsConcurrent Replication of Parallel and Distributed Simulations
Concurrent Replication of Parallel and Distributed SimulationsGabriele D'Angelo
 
Globecom 2015: Adaptive Raptor Carousel for 802.11
Globecom 2015: Adaptive Raptor Carousel for 802.11Globecom 2015: Adaptive Raptor Carousel for 802.11
Globecom 2015: Adaptive Raptor Carousel for 802.11Andrew Nix
 

Similar to Dynamic Adaptive Point Cloud Streaming (20)

A methodology for full system power modeling in heterogeneous data centers
A methodology for full system power modeling in  heterogeneous data centersA methodology for full system power modeling in  heterogeneous data centers
A methodology for full system power modeling in heterogeneous data centers
 
RECAP: The Simulation Approach
RECAP: The Simulation ApproachRECAP: The Simulation Approach
RECAP: The Simulation Approach
 
[EWiLi2016] Enabling power-awareness for the Xen Hypervisor
[EWiLi2016] Enabling power-awareness for the Xen Hypervisor[EWiLi2016] Enabling power-awareness for the Xen Hypervisor
[EWiLi2016] Enabling power-awareness for the Xen Hypervisor
 
GPU Renderfarm with Integrated Asset Management & Production System (AMPS)
GPU Renderfarm with Integrated Asset Management & Production System (AMPS)GPU Renderfarm with Integrated Asset Management & Production System (AMPS)
GPU Renderfarm with Integrated Asset Management & Production System (AMPS)
 
Multi Processor Architecture for image processing
Multi Processor Architecture for image processingMulti Processor Architecture for image processing
Multi Processor Architecture for image processing
 
Application Profiling at the HPCAC High Performance Center
Application Profiling at the HPCAC High Performance CenterApplication Profiling at the HPCAC High Performance Center
Application Profiling at the HPCAC High Performance Center
 
APSys Presentation Final copy2
APSys Presentation Final copy2APSys Presentation Final copy2
APSys Presentation Final copy2
 
RAMSES: Robust Analytic Models for Science at Extreme Scales
RAMSES: Robust Analytic Models for Science at Extreme ScalesRAMSES: Robust Analytic Models for Science at Extreme Scales
RAMSES: Robust Analytic Models for Science at Extreme Scales
 
CAD theory presentation.pptx .
CAD theory presentation.pptx                .CAD theory presentation.pptx                .
CAD theory presentation.pptx .
 
Three Challenges in Reliable Data Transport over Heterogeneous ...
Three Challenges in Reliable Data Transport over Heterogeneous ...Three Challenges in Reliable Data Transport over Heterogeneous ...
Three Challenges in Reliable Data Transport over Heterogeneous ...
 
Software-Defined Systems for Network-Aware Service Composition and Workflow P...
Software-Defined Systems for Network-Aware Service Composition and Workflow P...Software-Defined Systems for Network-Aware Service Composition and Workflow P...
Software-Defined Systems for Network-Aware Service Composition and Workflow P...
 
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degree
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degreeThe UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degree
The UCLouvain Public Defense of my EMJD-DC Double Doctorate Ph.D. degree
 
Standardising the compressed representation of neural networks
Standardising the compressed representation of neural networksStandardising the compressed representation of neural networks
Standardising the compressed representation of neural networks
 
Mirabilis_Design AMD Versal System-Level IP Library
Mirabilis_Design AMD Versal System-Level IP LibraryMirabilis_Design AMD Versal System-Level IP Library
Mirabilis_Design AMD Versal System-Level IP Library
 
byteLAKE's expertise across NVIDIA architectures and configurations
byteLAKE's expertise across NVIDIA architectures and configurationsbyteLAKE's expertise across NVIDIA architectures and configurations
byteLAKE's expertise across NVIDIA architectures and configurations
 
XLcloud 3-d remote rendering
XLcloud 3-d remote renderingXLcloud 3-d remote rendering
XLcloud 3-d remote rendering
 
Concurrent Replication of Parallel and Distributed Simulations
Concurrent Replication of Parallel and Distributed SimulationsConcurrent Replication of Parallel and Distributed Simulations
Concurrent Replication of Parallel and Distributed Simulations
 
Mantle for Developers
Mantle for DevelopersMantle for Developers
Mantle for Developers
 
SparkNet presentation
SparkNet presentationSparkNet presentation
SparkNet presentation
 
Globecom 2015: Adaptive Raptor Carousel for 802.11
Globecom 2015: Adaptive Raptor Carousel for 802.11Globecom 2015: Adaptive Raptor Carousel for 802.11
Globecom 2015: Adaptive Raptor Carousel for 802.11
 

More from Alpen-Adria-Universität

VEED: Video Encoding Energy and CO2 Emissions Dataset for AWS EC2 instances
VEED: Video Encoding Energy and CO2 Emissions Dataset for AWS EC2 instancesVEED: Video Encoding Energy and CO2 Emissions Dataset for AWS EC2 instances
VEED: Video Encoding Energy and CO2 Emissions Dataset for AWS EC2 instancesAlpen-Adria-Universität
 
GREEM: An Open-Source Energy Measurement Tool for Video Processing
GREEM: An Open-Source Energy Measurement Tool for Video ProcessingGREEM: An Open-Source Energy Measurement Tool for Video Processing
GREEM: An Open-Source Energy Measurement Tool for Video ProcessingAlpen-Adria-Universität
 
Optimal Quality and Efficiency in Adaptive Live Streaming with JND-Aware Low ...
Optimal Quality and Efficiency in Adaptive Live Streaming with JND-Aware Low ...Optimal Quality and Efficiency in Adaptive Live Streaming with JND-Aware Low ...
Optimal Quality and Efficiency in Adaptive Live Streaming with JND-Aware Low ...Alpen-Adria-Universität
 
VEEP: Video Encoding Energy and CO₂ Emission Prediction
VEEP: Video Encoding Energy and CO₂ Emission PredictionVEEP: Video Encoding Energy and CO₂ Emission Prediction
VEEP: Video Encoding Energy and CO₂ Emission PredictionAlpen-Adria-Universität
 
Content-adaptive Video Coding for HTTP Adaptive Streaming
Content-adaptive Video Coding for HTTP Adaptive StreamingContent-adaptive Video Coding for HTTP Adaptive Streaming
Content-adaptive Video Coding for HTTP Adaptive StreamingAlpen-Adria-Universität
 
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Video...
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Video...Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Video...
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Video...Alpen-Adria-Universität
 
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Vid...
Empowerment of Atypical Viewers  via Low-Effort Personalized Modeling  of Vid...Empowerment of Atypical Viewers  via Low-Effort Personalized Modeling  of Vid...
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Vid...Alpen-Adria-Universität
 
Optimizing Video Streaming for Sustainability and Quality: The Role of Prese...
Optimizing Video Streaming  for Sustainability and Quality: The Role of Prese...Optimizing Video Streaming  for Sustainability and Quality: The Role of Prese...
Optimizing Video Streaming for Sustainability and Quality: The Role of Prese...Alpen-Adria-Universität
 
Energy-Efficient Multi-Codec Bitrate-Ladder Estimation for Adaptive Video Str...
Energy-Efficient Multi-Codec Bitrate-Ladder Estimation for Adaptive Video Str...Energy-Efficient Multi-Codec Bitrate-Ladder Estimation for Adaptive Video Str...
Energy-Efficient Multi-Codec Bitrate-Ladder Estimation for Adaptive Video Str...Alpen-Adria-Universität
 
Machine Learning Based Resource Utilization Prediction in the Computing Conti...
Machine Learning Based Resource Utilization Prediction in the Computing Conti...Machine Learning Based Resource Utilization Prediction in the Computing Conti...
Machine Learning Based Resource Utilization Prediction in the Computing Conti...Alpen-Adria-Universität
 
Evaluation of Quality of Experience of ABR Schemes in Gaming Stream
Evaluation of Quality of Experience of ABR Schemes in Gaming StreamEvaluation of Quality of Experience of ABR Schemes in Gaming Stream
Evaluation of Quality of Experience of ABR Schemes in Gaming StreamAlpen-Adria-Universität
 
Network-Assisted Delivery of Adaptive Video Streaming Services through CDN, S...
Network-Assisted Delivery of Adaptive Video Streaming Services through CDN, S...Network-Assisted Delivery of Adaptive Video Streaming Services through CDN, S...
Network-Assisted Delivery of Adaptive Video Streaming Services through CDN, S...Alpen-Adria-Universität
 
Multi-access Edge Computing for Adaptive Video Streaming
Multi-access Edge Computing for Adaptive Video StreamingMulti-access Edge Computing for Adaptive Video Streaming
Multi-access Edge Computing for Adaptive Video StreamingAlpen-Adria-Universität
 
Policy-Driven Dynamic HTTP Adaptive Streaming Player Environment
Policy-Driven Dynamic HTTP Adaptive Streaming Player EnvironmentPolicy-Driven Dynamic HTTP Adaptive Streaming Player Environment
Policy-Driven Dynamic HTTP Adaptive Streaming Player EnvironmentAlpen-Adria-Universität
 
VE-Match: Video Encoding Matching-based Model for Cloud and Edge Computing In...
VE-Match: Video Encoding Matching-based Model for Cloud and Edge Computing In...VE-Match: Video Encoding Matching-based Model for Cloud and Edge Computing In...
VE-Match: Video Encoding Matching-based Model for Cloud and Edge Computing In...Alpen-Adria-Universität
 
Energy Consumption in Video Streaming: Components, Measurements, and Strategies
Energy Consumption in Video Streaming: Components, Measurements, and StrategiesEnergy Consumption in Video Streaming: Components, Measurements, and Strategies
Energy Consumption in Video Streaming: Components, Measurements, and StrategiesAlpen-Adria-Universität
 
Exploring the Energy Consumption of Video Streaming: Components, Challenges, ...
Exploring the Energy Consumption of Video Streaming: Components, Challenges, ...Exploring the Energy Consumption of Video Streaming: Components, Challenges, ...
Exploring the Energy Consumption of Video Streaming: Components, Challenges, ...Alpen-Adria-Universität
 
Video Coding Enhancements for HTTP Adaptive Streaming Using Machine Learning
Video Coding Enhancements for HTTP Adaptive Streaming Using Machine LearningVideo Coding Enhancements for HTTP Adaptive Streaming Using Machine Learning
Video Coding Enhancements for HTTP Adaptive Streaming Using Machine LearningAlpen-Adria-Universität
 
Optimizing QoE and Latency of Live Video Streaming Using Edge Computing a...
Optimizing  QoE and Latency of  Live Video Streaming Using  Edge Computing  a...Optimizing  QoE and Latency of  Live Video Streaming Using  Edge Computing  a...
Optimizing QoE and Latency of Live Video Streaming Using Edge Computing a...Alpen-Adria-Universität
 
SARENA: SFC-Enabled Architecture for Adaptive Video Streaming Applications
SARENA: SFC-Enabled Architecture for Adaptive Video Streaming ApplicationsSARENA: SFC-Enabled Architecture for Adaptive Video Streaming Applications
SARENA: SFC-Enabled Architecture for Adaptive Video Streaming ApplicationsAlpen-Adria-Universität
 

More from Alpen-Adria-Universität (20)

VEED: Video Encoding Energy and CO2 Emissions Dataset for AWS EC2 instances
VEED: Video Encoding Energy and CO2 Emissions Dataset for AWS EC2 instancesVEED: Video Encoding Energy and CO2 Emissions Dataset for AWS EC2 instances
VEED: Video Encoding Energy and CO2 Emissions Dataset for AWS EC2 instances
 
GREEM: An Open-Source Energy Measurement Tool for Video Processing
GREEM: An Open-Source Energy Measurement Tool for Video ProcessingGREEM: An Open-Source Energy Measurement Tool for Video Processing
GREEM: An Open-Source Energy Measurement Tool for Video Processing
 
Optimal Quality and Efficiency in Adaptive Live Streaming with JND-Aware Low ...
Optimal Quality and Efficiency in Adaptive Live Streaming with JND-Aware Low ...Optimal Quality and Efficiency in Adaptive Live Streaming with JND-Aware Low ...
Optimal Quality and Efficiency in Adaptive Live Streaming with JND-Aware Low ...
 
VEEP: Video Encoding Energy and CO₂ Emission Prediction
VEEP: Video Encoding Energy and CO₂ Emission PredictionVEEP: Video Encoding Energy and CO₂ Emission Prediction
VEEP: Video Encoding Energy and CO₂ Emission Prediction
 
Content-adaptive Video Coding for HTTP Adaptive Streaming
Content-adaptive Video Coding for HTTP Adaptive StreamingContent-adaptive Video Coding for HTTP Adaptive Streaming
Content-adaptive Video Coding for HTTP Adaptive Streaming
 
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Video...
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Video...Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Video...
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Video...
 
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Vid...
Empowerment of Atypical Viewers  via Low-Effort Personalized Modeling  of Vid...Empowerment of Atypical Viewers  via Low-Effort Personalized Modeling  of Vid...
Empowerment of Atypical Viewers via Low-Effort Personalized Modeling of Vid...
 
Optimizing Video Streaming for Sustainability and Quality: The Role of Prese...
Optimizing Video Streaming  for Sustainability and Quality: The Role of Prese...Optimizing Video Streaming  for Sustainability and Quality: The Role of Prese...
Optimizing Video Streaming for Sustainability and Quality: The Role of Prese...
 
Energy-Efficient Multi-Codec Bitrate-Ladder Estimation for Adaptive Video Str...
Energy-Efficient Multi-Codec Bitrate-Ladder Estimation for Adaptive Video Str...Energy-Efficient Multi-Codec Bitrate-Ladder Estimation for Adaptive Video Str...
Energy-Efficient Multi-Codec Bitrate-Ladder Estimation for Adaptive Video Str...
 
Machine Learning Based Resource Utilization Prediction in the Computing Conti...
Machine Learning Based Resource Utilization Prediction in the Computing Conti...Machine Learning Based Resource Utilization Prediction in the Computing Conti...
Machine Learning Based Resource Utilization Prediction in the Computing Conti...
 
Evaluation of Quality of Experience of ABR Schemes in Gaming Stream
Evaluation of Quality of Experience of ABR Schemes in Gaming StreamEvaluation of Quality of Experience of ABR Schemes in Gaming Stream
Evaluation of Quality of Experience of ABR Schemes in Gaming Stream
 
Network-Assisted Delivery of Adaptive Video Streaming Services through CDN, S...
Network-Assisted Delivery of Adaptive Video Streaming Services through CDN, S...Network-Assisted Delivery of Adaptive Video Streaming Services through CDN, S...
Network-Assisted Delivery of Adaptive Video Streaming Services through CDN, S...
 
Multi-access Edge Computing for Adaptive Video Streaming
Multi-access Edge Computing for Adaptive Video StreamingMulti-access Edge Computing for Adaptive Video Streaming
Multi-access Edge Computing for Adaptive Video Streaming
 
Policy-Driven Dynamic HTTP Adaptive Streaming Player Environment
Policy-Driven Dynamic HTTP Adaptive Streaming Player EnvironmentPolicy-Driven Dynamic HTTP Adaptive Streaming Player Environment
Policy-Driven Dynamic HTTP Adaptive Streaming Player Environment
 
VE-Match: Video Encoding Matching-based Model for Cloud and Edge Computing In...
VE-Match: Video Encoding Matching-based Model for Cloud and Edge Computing In...VE-Match: Video Encoding Matching-based Model for Cloud and Edge Computing In...
VE-Match: Video Encoding Matching-based Model for Cloud and Edge Computing In...
 
Energy Consumption in Video Streaming: Components, Measurements, and Strategies
Energy Consumption in Video Streaming: Components, Measurements, and StrategiesEnergy Consumption in Video Streaming: Components, Measurements, and Strategies
Energy Consumption in Video Streaming: Components, Measurements, and Strategies
 
Exploring the Energy Consumption of Video Streaming: Components, Challenges, ...
Exploring the Energy Consumption of Video Streaming: Components, Challenges, ...Exploring the Energy Consumption of Video Streaming: Components, Challenges, ...
Exploring the Energy Consumption of Video Streaming: Components, Challenges, ...
 
Video Coding Enhancements for HTTP Adaptive Streaming Using Machine Learning
Video Coding Enhancements for HTTP Adaptive Streaming Using Machine LearningVideo Coding Enhancements for HTTP Adaptive Streaming Using Machine Learning
Video Coding Enhancements for HTTP Adaptive Streaming Using Machine Learning
 
Optimizing QoE and Latency of Live Video Streaming Using Edge Computing a...
Optimizing  QoE and Latency of  Live Video Streaming Using  Edge Computing  a...Optimizing  QoE and Latency of  Live Video Streaming Using  Edge Computing  a...
Optimizing QoE and Latency of Live Video Streaming Using Edge Computing a...
 
SARENA: SFC-Enabled Architecture for Adaptive Video Streaming Applications
SARENA: SFC-Enabled Architecture for Adaptive Video Streaming ApplicationsSARENA: SFC-Enabled Architecture for Adaptive Video Streaming Applications
SARENA: SFC-Enabled Architecture for Adaptive Video Streaming Applications
 

Recently uploaded

From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationSafe Software
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxOnBoard
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking MenDelhi Call girls
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreternaman860154
 
Azure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAzure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAndikSusilo4
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptxHampshireHUG
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Paola De la Torre
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationSafe Software
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsEnterprise Knowledge
 
How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?XfilesPro
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...HostedbyConfluent
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxKatpro Technologies
 
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphSIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphNeo4j
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationRadu Cotescu
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsMaria Levchenko
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerThousandEyes
 

Recently uploaded (20)

From Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time AutomationFrom Event to Action: Accelerate Your Decision Making with Real-Time Automation
From Event to Action: Accelerate Your Decision Making with Real-Time Automation
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
Transcript: #StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
Maximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptxMaximizing Board Effectiveness 2024 Webinar.pptx
Maximizing Board Effectiveness 2024 Webinar.pptx
 
08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men08448380779 Call Girls In Civil Lines Women Seeking Men
08448380779 Call Girls In Civil Lines Women Seeking Men
 
Presentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreterPresentation on how to chat with PDF using ChatGPT code interpreter
Presentation on how to chat with PDF using ChatGPT code interpreter
 
Azure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & ApplicationAzure Monitor & Application Insight to monitor Infrastructure & Application
Azure Monitor & Application Insight to monitor Infrastructure & Application
 
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
Neo4j - How KGs are shaping the future of Generative AI at AWS Summit London ...
 
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
04-2024-HHUG-Sales-and-Marketing-Alignment.pptx
 
Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101Salesforce Community Group Quito, Salesforce 101
Salesforce Community Group Quito, Salesforce 101
 
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry InnovationBeyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
Beyond Boundaries: Leveraging No-Code Solutions for Industry Innovation
 
IAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI SolutionsIAC 2024 - IA Fast Track to Search Focused AI Solutions
IAC 2024 - IA Fast Track to Search Focused AI Solutions
 
How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?How to Remove Document Management Hurdles with X-Docs?
How to Remove Document Management Hurdles with X-Docs?
 
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
Transforming Data Streams with Kafka Connect: An Introduction to Single Messa...
 
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptxFactors to Consider When Choosing Accounts Payable Services Providers.pptx
Factors to Consider When Choosing Accounts Payable Services Providers.pptx
 
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge GraphSIEMENS: RAPUNZEL – A Tale About Knowledge Graph
SIEMENS: RAPUNZEL – A Tale About Knowledge Graph
 
Scaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organizationScaling API-first – The story of a global engineering organization
Scaling API-first – The story of a global engineering organization
 
Handwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed textsHandwritten Text Recognition for manuscripts and early printed texts
Handwritten Text Recognition for manuscripts and early printed texts
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 

Dynamic Adaptive Point Cloud Streaming

  • 1. Dynamic Adaptive Point Cloud Streaming ACM Packet Video 2018 (PV’18) Mohammad Hosseini University of Illinois at Urbana-Champaign (UIUC) shossen2@Illinois.edu Christian Timmerer Alpen-Adria-Universitat Klagenfurt, Bitmovin Inc. christian.timmerer@itec.uni-klu.ac.at
  • 2. 2 /28 Introduction • 3D point clouds are set of points represented in the 3D space • Recently emerged as an advanced representation of immersive media • Can be captured using camera arrays and depth sensors in various setups
  • 3. 3 /28 • Point clouds are very bulky! – Can consist of billions of points – Highly resource intensive – The data transmission rate in raw point-cloud frames in an application with 30 fps can be as high as 6 Gbps Motivation Goal: How to achieve a balance between the streaming quality and the available resources on the display devices without much negative on user’s viewing experience
  • 4. 4 /28 Related Work • Dynamic Adaptive Streaming over HTTP (MPEG-DASH) – A client chooses a video segment with the appropriate quality (bit rate, resolution, etc.) based on its constrained resources such as bandwidth – The video content is stored on an HTTP server, and is accompanied by a Media Presentation Description (MPD) – We extend the semantics of MPEG-DASH towards a dynamic point cloud streaming system • To enable view-aware and bandwidth-aware adaptation through transmitting different point cloud representations relative to the user’s resources • Point Cloud Compression (PCC) – Existing works are mostly centered around point cloud compression using geometry-based approaches – On the MPEG side, a new work item has been initiated for Point Cloud Compression, or in short, MPEG PCC to cover topics in lossy and lossless point cloud compression
  • 5. 5 /28 • DASH-PC: Extend DASH towards adaptive point cloud streaming • Point-cloud density reduction based on distance-performance-quality trade-offs Proposed Approach 15K points 30K points 150K points 1M points Storage, Processing, Power, etc.
  • 6. 6 /28 • Quality adaptation based on distance • Human Visual Acuity! – 60 Pixel per degree for adults Proposed Approach D … distance of a user from the display screen S … scale factor of a point cloud object D’ … distance of the camera position from the centroid of the object’s bounding box
  • 7. 7 /28 Proposed Approach Low Density (Rep. 3) Medium Density (Rep. 2) High Density (Rep. 1)
  • 9. 9 /28 • Similar to DASH, the adaptation feature of DASH-PC is described by an XML-formatted manifest containing metadata • Hierarchical manifest, divided into separate PC frames. Each frame including a variety of adaptation sets: – Multiple quality alternatives – Index of each frame – The frame’s HTTP location – Level of Density (LoD) representations DASH-PC: Manifest
  • 10. 10 /28 • An example manifest DASH-PC: Manifest
  • 11. 11 /28 • Re-use some of the DASH mime types for a better compliance with MPEG-DASH – @format: specifies the point cloud container format (e.g., ply) – @encoding: specifies the point cloud encoding (ASCII or binary) – @frames: specifies the number of point cloud frames followed – @type: shows if the manifest may be updated during a session (’static’ (default) or ’dynamic’) – BaseURL: specifies the base HTTP URL of the point cloud DASH-PC: Manifest
  • 12. 12 /28 • Frame – Typically represents a single point cloud model – For each frame, a consistent set of multiple adaptation sets of that point cloud model are available – @id: a unique identifier of the frame – BaseURL: specifies the base HTTP URL of the frame – AdaptationSet: specifies the sub-clause adaptation sets DASH-PC: Manifest
  • 13. 13 /28 • AdaptationSet: – Consists of multiple interchangeable quality versions of a point cloud frame – It can carry additional attributes such as: • Density sub-sampling methods • Data structure used • Compression techniques • Viewport information – @id: a unique identifier of the adaptation set – BaseURL: specifies the base HTTP URL of the adaptation set – Representation: specifies the sub-clause representations DASH-PC: Manifest
  • 14. 14 /28 • Representation – A single point cloud representation is sufficient for rendering and visualization of a model – A client can switch from one representation to another at any time • To adapt to its varying resources and parameters such as network bandwidth, energy budget, visual preferences, etc. – @id: unique identifier of the representation – BaseURL: specifies the base HTTP URL of the representation – @density: specifies the density of the point cloud model in terms of the number of points – @size: specifies the size of the point cloud model in terms of the storage volume – Segment: specifies the sub-clause segments DASH-PC: Manifest
  • 15. 15 /28 • Segment – Describe sub-models based on spatial segmentation which can be used to retrieve partial point cloud models – For better compliance with the features of viewport adaptations – @id: unique identifier of the adaptation set – BaseURL: specifies the HTTP URL of the segment – @density: specifies the density of the point cloud model in terms of the number of points – @size: specifies the size of the point cloud model in terms of the storage volume DASH-PC: Manifest
  • 16. 16 /28 • Proposed three different point-cloud density reduction algorithms • Based on the concept of spatial clustering of neighbor points in the 3D space and sampling points in each cluster Density Reduction
  • 17. 17 /28 • Algorithm 1: – Clusters are simply constructed through points geometrically being sorted in the 3D space – Each cluster with near-minimum spatial Euclidean distance of consisting points • Algorithm 2: – Use the notion of histogram, and construct a data structure called “low-resolution point cloud density tree” • Algorithm 3: – Enhanced version of Algorithm 2 Density Reduction: Algorithms Overview
  • 18. 18 /28 • Developed tools and simple APIs for users to construct representations based on sub-sampling and scaling • subsample(int percentage, String inputpc, bool isIterative) – Generates a sub-sampled point cloud model with the percentage integer specifying the density of the output point cloud model relative to the input point cloud – isIterative flag specifies if the process is iterated over multiple successive frames • scale(int percentage, String inputpc, bool isIterative) – Generates a scaled point cloud model given the scaling ratio • optimize(int distance, int scale, int inputpc, bool isIterative) – Automatically generates a sub-sampled point cloud with optimum PPI density given the distance (in inch) and/or the scale Density Reduction: User API
  • 19. 19 /28 • Used Point Cloud Library (PCL 1.8.1) to develop a dynamic point cloud visualization and streaming prototype • Used four different sequences provided by JPEG Pleno Database containing 8i voxelized dynamic point cloud sequences – longdress, loot, redandblack, and soldier – In each sequence, the full object is captured by 42 RGB cameras at 30 fps, over a 10 second period Experimental Setup
  • 20. 20 /28 • Intel Xeon E5520 64-bit machine with quad-core 2.27 GHz CPU, 12 GB RAM, and Gallium 0.4 on AMD GPU • Ubuntu 16.04 LTS • Server-client setup – Used our machine as the local HTTP streaming server to filter the negative impacts of network latency and bandwidth variations on the experimental results – To focus purely on the amount of bandwidth savings, rendering performance, and objective quality measurement Experimental Setup
  • 21. 21 /28 • Applied four sub-sampling ratios to the test sequences, and prepared 5 different representations in total for each point cloud frame – REP1 to REP5 representing highest to lowest density, given sub-sampling ratios of 1 to 5 • Measured three different factors: – Total bandwidth usage  directly correlated to the sub-sampling ratio • Total bandwidth saving of approximately 50% and 80% was achieved in raw point cloud frames given sub-sampling ratios of 2 and 5, respectively – Rendering performance (FPS) – Objective quality (PSNR) • Compared our results with the baseline scenario where no adaptation is employed • Each trial of our experiment was run for a total of 30 consecutive test sequences • Repeated each experiment 6 times – To ensure that the standard deviations for variable metrics are within acceptable limits Evaluation
  • 22. 22 /28 • Measured the client’s rendering performance in terms of the average FPS – Measured FPS continuously for 100 times during each session • Recorded the data when adaptations are applied – Lower representations (REP2 and REP5 with sub-sample ratios of 2 and 5) compared to the baseline case where no adaptation is applied (REP1) Evaluation: Rendering Performance
  • 23. 23 /28 • Rendering Performance (FPS) – Results for only a subset of our experiments on all of our benchmarks under Algorithm 1 – Our adaptations can significantly improve the FPS up to more than 10x Evaluation: Rendering Performance
  • 24. 24 /28 • Objective Quality (PSNR) – Extensive measurements on the objective quality using PSNR of point-to-point distortions to verify our assumptions for visual acuity – Used the official MPEG PCC quality metric software • Compares an original point cloud with an adapted model and provides numerical values for point cloud PSNR – Averaged the PSNR measurements over all dynamic point cloud frames, and collected the experimental results Evaluation: Objective Quality
  • 25. 25 /28 • Objective Quality (PSNR) – PSNR under 2x, 4x, and 8x scale-down ratios – Confirms our earlier assumptions in regards to the effects of scaling point clouds on the human’s visual acuity Evaluation: Objective Quality
  • 26. 26 /28 • Visual View – Illustrating three consecutive frames from highest representation (REP1) (left) to the lowest representation (REP5) (right) Evaluation: Visual View Visual comparison of a specific frame within Red&Black under 3 sample representations: (left) REP1, (middle) REP2, (right) REP5.
  • 27. 27 /28 Conclusion and Future Work • Conclusions: – Proposed DASH-PC, a dynamic point cloud streaming adaptation system to tackle the high bandwidth and processing requirements of dense point clouds – Studied the semantic link of adaptive streaming with a user’s visual acuity, limited bandwidth, and performance overhead – Initial experimental results show that our adaptations can significantly save bandwidth and improve rendering performance on point cloud streaming with negligible noticeable quality impacts • Future Work: – Study various rate adaptation algorithms and evaluate their quality impacts

Editor's Notes

  1. PPI … pixel per inch
  2. With this negligible noticeable quality impacts, sometimes not even perceptible, while it maintains possibly highest quality to ensure a satisfactory user experience. Overall, considering the significant bandwidth saving, performance improvement, and lower hardware exhaustion achieved using our adaptations, it is reasonable to believe that many would accept our adaptations to respect the valuable limited resources in an immersive environment.
  3. Email the first author with questions?