SlideShare a Scribd company logo
Subjective Quality Assessment in
Stereoscopic Video Based on
Analyzing Parallax and Disparity
Juan Pedro López Velasco, Juan Antonio Rodrigo,
David Jiménez and José Manuel Menéndez
Universidad Politécnica de Madrid
Las Vegas NV, 10th January 2015
Index
• Problem description: Visual discomfort in 3D
• Work methodology:
 Characterization of individual sequences
 Description of pairs of sequences
 Subjective quality assessment
• Results and examples
• Future work
• Conclusions
Problem description (I)
• The differences and variations in 3D Disparity
may cause visual discomfort.
• What is disparity in 3D video?
• Disparity may offer an incredible experience.
• The eye focus the objects. Accommodation
of the eyes needs enough time to adapt to
changes for correct vision of 3D videos.
Problem description (II)
Problem description (III)
• If video sequences are not well controlled,
especially in abrupt changes, which need fast
converging process in the eye…
What happens??
Problem description (IV)
• The eye has not enough time to focus
objects, discomfort, annoyance and,
consequently, headache may occur.
• Because convergence in 3D sequences is
more complex than when contemplating
bidimmensional video.
Problem description (V)
• But…
– How do we evaluate the reactions of
human eye derived from abrupt changes
when contemplating 3D video?
• Answer:
– With subjective assessment showing the
spectator pairs of sequences and analyzing the
abrupt changes among us.
– Knowing the variations between parallax
diagrams is useful for this effect.
Problem description (VI)
Examples of 3D disparity (I)
Examples of 3D disparity (II)
Work methodology
Characterization of
individual video sequences
Sequence
1
Sequence
2
Combination of video
pair sequences
Sequence
Motion
Parallax
Distribution
1
2
High variety of types of
transitions
Subjective assessment
with pairs of sequences
for transition analysis
3
Analysis of when visual
discomfort happens4
Characterization of individual
video sequences
• Parallax diagrams
– Ranges and variance of negative and
positive parallax
• Distribution of parallax in the image.
• Values of Temporal (TI) and Spatial (SI)
Information (also in 2D tests)
– Motion (TI)
– Spatial complexity and detail (SI)
Parallax and disparity diagram
Visual distribution
of parallax
Parallax
Histograms
With the use
of SAD
techniques
Combination of video pair
sequences
– Analysis of types of changes or transitions
between two 3D sequences.
– Types of sequences transitions:
• Combination of sequences with positive and
negative parallax
• Combination of sequences with higher or lower
variance of negative/positive parallax
• Combination of sequences with objects located
in different / same regions of the image
Subjective assessment
• Battery of tests with 3D high-definition
video sequences
• Annoyance 5-notes Scale for evaluating
the transition between each pair of videos
MOS
Scale
Annoyance derived from
transition
Quality of Experience
5 Very comfortable Excellent Experience
4 Comfortable Good Experience
3 Mildly uncomfortable No visual discomfort
2 Uncomfortable Visual discomfort
1 Extremely uncomfortable High visual discomfort
Device Philips 65PFL9708S
Resolution 3840 x 2160p
Brightness 450 cd/m²
Size 65” Widescreen
Technology LED Ultra HD
3D Passive glasses
Example 1
• Transition 12: “Angel” to “Ladder”
 40% of the people give a score that manifests visual
discomfort
Example 2
• Transition 14: “Spaceship” to
“Astronaut”
 Negative parallax in right side of first video to
negative/positive combination
Example 3
• Transition 16: “Station” to “Itaca3d”
 This is the worst scored transition in the tests
↑↑Motion
↑↑Motion
Hiperstereoscopy!
Example 4
• Transition 2: “Boxers” to “Dance”
 Negative parallax located in different areas, less
annoyance for observers.
Example 5
• Transition 4: “Hall” to “Laboratory”
 Both videos with negative parallax in both videos and
window violation → low scores. Window violation!
Conclusions (I)
• Motion is a key element of production of
visual discomfort, but it is not the only one.
When it is combined with sequences with
high negative parallax may generate visual
discomfort.
• If the negative parallax in hyperstereoscopy
is located in different parts of the sequences
in the transition, visual discomfort appears
less than when located in same areas.
Conclusions (II)
• Fast variation of negative parallax is usually
the main source of visual discomfort,
especially when the transition is produced to
a content with a completely different disparity
diagram.
• Only hyperstereoscopy (i.e. pixels with
negative parallax with disparities higher than
5) in the sequence is not enough for
detecting visual discomfort, it is the transition
what provokes the discomfort.
Future work
• After detecting the main sources of
visual discomfort…
Developing recommendations and
guidelines for 3D contents creators.
Generating tools for automatic
detection of discomfort in 3D videos.
Thanks for your attention!!

More Related Content

PDF
Tipo de empresa
PDF
Operação Mercúrio
PPTX
Workshop abrindo a caixa de pandora
RTF
Documento 14
PPTX
Academia MOLTENIS
PDF
Novo apendice ii__enderecos_2013-10-23_00_08_35
DOCX
Practican6 listado de palabras
PDF
Hddumanual control mixer
Tipo de empresa
Operação Mercúrio
Workshop abrindo a caixa de pandora
Documento 14
Academia MOLTENIS
Novo apendice ii__enderecos_2013-10-23_00_08_35
Practican6 listado de palabras
Hddumanual control mixer

Similar to Subjective Quality Assessment in Stereoscopic Video Based on Analyzing Parallax and Disparity (20)

PPTX
Presentación Tesis 08022016
PDF
The Effectiveness of 2D-3D Converters in Rendering Natural Water Phenomena
PDF
Sony
PPTX
Tele immersion
DOCX
PDF
Virtual viewpoint three dimensional panorama
PDF
VIRTUAL VIEWPOINT THREE-DIMENSIONAL PANORAMA
PDF
VIRTUAL VIEWPOINT THREE-DIMENSIONAL PANORAMA
PDF
VIRTUAL VIEWPOINT THREE-DIMENSIONAL PANORAMA
PDF
8. 10179 13825-1-ed edit iqbal
PDF
3D Subjective Testing
PDF
Phase-Modulation Based 3D technology
DOC
Quality assessment of stereoscopic 3 d image compression by binocular integra...
PDF
Stevenson Eye Tracking With The Adaptive Optics Scanning Laser Ophthalmoscope
PPTX
Color-plus-Depth Level-of-Detail in 3D Tele-immersive Video: A Psychophysical...
PDF
Insertion of Impairments in Test Video Sequences for Quality Assessment Based...
PDF
Mulligan Robust Optical Eye Detection During Head Movement
PPT
Visual thinking colin_ware_lectures_2013_7_3_d space perception
PDF
Synthesizing pseudo 2.5 d content from monocular videos for mixed reality
Presentación Tesis 08022016
The Effectiveness of 2D-3D Converters in Rendering Natural Water Phenomena
Sony
Tele immersion
Virtual viewpoint three dimensional panorama
VIRTUAL VIEWPOINT THREE-DIMENSIONAL PANORAMA
VIRTUAL VIEWPOINT THREE-DIMENSIONAL PANORAMA
VIRTUAL VIEWPOINT THREE-DIMENSIONAL PANORAMA
8. 10179 13825-1-ed edit iqbal
3D Subjective Testing
Phase-Modulation Based 3D technology
Quality assessment of stereoscopic 3 d image compression by binocular integra...
Stevenson Eye Tracking With The Adaptive Optics Scanning Laser Ophthalmoscope
Color-plus-Depth Level-of-Detail in 3D Tele-immersive Video: A Psychophysical...
Insertion of Impairments in Test Video Sequences for Quality Assessment Based...
Mulligan Robust Optical Eye Detection During Head Movement
Visual thinking colin_ware_lectures_2013_7_3_d space perception
Synthesizing pseudo 2.5 d content from monocular videos for mixed reality
Ad

More from Universidad Politécnica de Madrid (13)

PPT
Presentation Crosshaul (Nem Summit, Madrid) - 29Nov2017
PPT
Presentation HbbTV (Nem Summit, Madrid) - 30Nov2017
PPT
Presentation Quality Probe (Vilamoura, Portugal, IADIS Applied Computing Conf...
PPTX
Medidas de calidad subjetiva y objetiva para vídeo UHD en el entorno del proy...
PPTX
El arte del vídeo musical y su influencia en la apreciación de la música
PPT
Empleo del software Autodesk Maya para modelado, animación, creación de efect...
PDF
La mujer a ojos de la vanguardia
PPTX
[EQSA] El sistema visual humano
PPT
Impact of new technologies and social network on a secondary education theatr...
PDF
Libreto "Child Soldier" de la Compañía Al-Satt Teatro Danza
PPT
No-Reference Algorithms for Video Quality Assessment based on Artifact Evalua...
PPTX
Lenguaje audiovisual: El cine digital y las nuevas tecnologías como soporte p...
PPTX
Lenguaje audiovisual: El cine digital y las nuevas tecnologías como soporte p...
Presentation Crosshaul (Nem Summit, Madrid) - 29Nov2017
Presentation HbbTV (Nem Summit, Madrid) - 30Nov2017
Presentation Quality Probe (Vilamoura, Portugal, IADIS Applied Computing Conf...
Medidas de calidad subjetiva y objetiva para vídeo UHD en el entorno del proy...
El arte del vídeo musical y su influencia en la apreciación de la música
Empleo del software Autodesk Maya para modelado, animación, creación de efect...
La mujer a ojos de la vanguardia
[EQSA] El sistema visual humano
Impact of new technologies and social network on a secondary education theatr...
Libreto "Child Soldier" de la Compañía Al-Satt Teatro Danza
No-Reference Algorithms for Video Quality Assessment based on Artifact Evalua...
Lenguaje audiovisual: El cine digital y las nuevas tecnologías como soporte p...
Lenguaje audiovisual: El cine digital y las nuevas tecnologías como soporte p...
Ad

Recently uploaded (20)

PPTX
20250228 LYD VKU AI Blended-Learning.pptx
PPTX
sap open course for s4hana steps from ECC to s4
PPTX
ACSFv1EN-58255 AWS Academy Cloud Security Foundations.pptx
PPTX
Detection-First SIEM: Rule Types, Dashboards, and Threat-Informed Strategy
PPTX
Programs and apps: productivity, graphics, security and other tools
PDF
cuic standard and advanced reporting.pdf
DOCX
The AUB Centre for AI in Media Proposal.docx
PPTX
Spectroscopy.pptx food analysis technology
PDF
NewMind AI Weekly Chronicles - August'25 Week I
PDF
Unlocking AI with Model Context Protocol (MCP)
PDF
The Rise and Fall of 3GPP – Time for a Sabbatical?
PPT
“AI and Expert System Decision Support & Business Intelligence Systems”
PPT
Teaching material agriculture food technology
PDF
Encapsulation_ Review paper, used for researhc scholars
PDF
7 ChatGPT Prompts to Help You Define Your Ideal Customer Profile.pdf
PDF
Per capita expenditure prediction using model stacking based on satellite ima...
PDF
TokAI - TikTok AI Agent : The First AI Application That Analyzes 10,000+ Vira...
PDF
Mobile App Security Testing_ A Comprehensive Guide.pdf
PDF
Spectral efficient network and resource selection model in 5G networks
PDF
Electronic commerce courselecture one. Pdf
20250228 LYD VKU AI Blended-Learning.pptx
sap open course for s4hana steps from ECC to s4
ACSFv1EN-58255 AWS Academy Cloud Security Foundations.pptx
Detection-First SIEM: Rule Types, Dashboards, and Threat-Informed Strategy
Programs and apps: productivity, graphics, security and other tools
cuic standard and advanced reporting.pdf
The AUB Centre for AI in Media Proposal.docx
Spectroscopy.pptx food analysis technology
NewMind AI Weekly Chronicles - August'25 Week I
Unlocking AI with Model Context Protocol (MCP)
The Rise and Fall of 3GPP – Time for a Sabbatical?
“AI and Expert System Decision Support & Business Intelligence Systems”
Teaching material agriculture food technology
Encapsulation_ Review paper, used for researhc scholars
7 ChatGPT Prompts to Help You Define Your Ideal Customer Profile.pdf
Per capita expenditure prediction using model stacking based on satellite ima...
TokAI - TikTok AI Agent : The First AI Application That Analyzes 10,000+ Vira...
Mobile App Security Testing_ A Comprehensive Guide.pdf
Spectral efficient network and resource selection model in 5G networks
Electronic commerce courselecture one. Pdf

Subjective Quality Assessment in Stereoscopic Video Based on Analyzing Parallax and Disparity

  • 1. Subjective Quality Assessment in Stereoscopic Video Based on Analyzing Parallax and Disparity Juan Pedro López Velasco, Juan Antonio Rodrigo, David Jiménez and José Manuel Menéndez Universidad Politécnica de Madrid Las Vegas NV, 10th January 2015
  • 2. Index • Problem description: Visual discomfort in 3D • Work methodology:  Characterization of individual sequences  Description of pairs of sequences  Subjective quality assessment • Results and examples • Future work • Conclusions
  • 3. Problem description (I) • The differences and variations in 3D Disparity may cause visual discomfort. • What is disparity in 3D video?
  • 4. • Disparity may offer an incredible experience. • The eye focus the objects. Accommodation of the eyes needs enough time to adapt to changes for correct vision of 3D videos. Problem description (II)
  • 5. Problem description (III) • If video sequences are not well controlled, especially in abrupt changes, which need fast converging process in the eye… What happens??
  • 6. Problem description (IV) • The eye has not enough time to focus objects, discomfort, annoyance and, consequently, headache may occur. • Because convergence in 3D sequences is more complex than when contemplating bidimmensional video.
  • 7. Problem description (V) • But… – How do we evaluate the reactions of human eye derived from abrupt changes when contemplating 3D video?
  • 8. • Answer: – With subjective assessment showing the spectator pairs of sequences and analyzing the abrupt changes among us. – Knowing the variations between parallax diagrams is useful for this effect. Problem description (VI)
  • 9. Examples of 3D disparity (I)
  • 10. Examples of 3D disparity (II)
  • 11. Work methodology Characterization of individual video sequences Sequence 1 Sequence 2 Combination of video pair sequences Sequence Motion Parallax Distribution 1 2 High variety of types of transitions Subjective assessment with pairs of sequences for transition analysis 3 Analysis of when visual discomfort happens4
  • 12. Characterization of individual video sequences • Parallax diagrams – Ranges and variance of negative and positive parallax • Distribution of parallax in the image. • Values of Temporal (TI) and Spatial (SI) Information (also in 2D tests) – Motion (TI) – Spatial complexity and detail (SI)
  • 13. Parallax and disparity diagram Visual distribution of parallax Parallax Histograms With the use of SAD techniques
  • 14. Combination of video pair sequences – Analysis of types of changes or transitions between two 3D sequences. – Types of sequences transitions: • Combination of sequences with positive and negative parallax • Combination of sequences with higher or lower variance of negative/positive parallax • Combination of sequences with objects located in different / same regions of the image
  • 15. Subjective assessment • Battery of tests with 3D high-definition video sequences • Annoyance 5-notes Scale for evaluating the transition between each pair of videos MOS Scale Annoyance derived from transition Quality of Experience 5 Very comfortable Excellent Experience 4 Comfortable Good Experience 3 Mildly uncomfortable No visual discomfort 2 Uncomfortable Visual discomfort 1 Extremely uncomfortable High visual discomfort Device Philips 65PFL9708S Resolution 3840 x 2160p Brightness 450 cd/m² Size 65” Widescreen Technology LED Ultra HD 3D Passive glasses
  • 16. Example 1 • Transition 12: “Angel” to “Ladder”  40% of the people give a score that manifests visual discomfort
  • 17. Example 2 • Transition 14: “Spaceship” to “Astronaut”  Negative parallax in right side of first video to negative/positive combination
  • 18. Example 3 • Transition 16: “Station” to “Itaca3d”  This is the worst scored transition in the tests ↑↑Motion ↑↑Motion Hiperstereoscopy!
  • 19. Example 4 • Transition 2: “Boxers” to “Dance”  Negative parallax located in different areas, less annoyance for observers.
  • 20. Example 5 • Transition 4: “Hall” to “Laboratory”  Both videos with negative parallax in both videos and window violation → low scores. Window violation!
  • 21. Conclusions (I) • Motion is a key element of production of visual discomfort, but it is not the only one. When it is combined with sequences with high negative parallax may generate visual discomfort. • If the negative parallax in hyperstereoscopy is located in different parts of the sequences in the transition, visual discomfort appears less than when located in same areas.
  • 22. Conclusions (II) • Fast variation of negative parallax is usually the main source of visual discomfort, especially when the transition is produced to a content with a completely different disparity diagram. • Only hyperstereoscopy (i.e. pixels with negative parallax with disparities higher than 5) in the sequence is not enough for detecting visual discomfort, it is the transition what provokes the discomfort.
  • 23. Future work • After detecting the main sources of visual discomfort… Developing recommendations and guidelines for 3D contents creators. Generating tools for automatic detection of discomfort in 3D videos.
  • 24. Thanks for your attention!!