Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Is there a relationship between Mean Opinion Score (MOS) and Just Noticeable Difference (JND)?

Jingwen Zhu, Hadi Amirpour, Wei Zhou, Patrick Le Callet

TL;DR

This study addresses how Mean Opinion Score (MOS) relates to Just Noticeable Difference (JND) in high-quality video regions, a key concern for perceptual bitrate ladders. It uses a Degradation Category Rating (DCR) protocol on data derived from the HD-VJND JND dataset, modeling JND distributions with Satisfied User Ratio (SUR) and examining MOS at the 75% SUR anchors of the 1st and 2nd JND. The results show MOS values near ideal targets, with the 1st JND anchor at 1080p yielding about $4.79$ and the 2nd JND decreasing with increased distortion, e.g., $MOS ightarrow 4.50$ at the ideal anchor; importantly, the mapping is one-way: MOS can inform the JND level but cannot uniquely determine it due to overlapping confidence intervals and limited study power. These findings have practical implications for perceptual bitrate ladder design and emphasize the need for larger-scale subjective studies to robustly connect MOS and JND in premium video experiences.

Abstract

Evaluating perceived video quality is essential for ensuring high Quality of Experience (QoE) in modern streaming applications. While existing subjective datasets and Video Quality Metrics (VQMs) cover a broad quality range, many practical use cases especially for premium users focus on high quality scenarios requiring finer granularity. Just Noticeable Difference (JND) has emerged as a key concept for modeling perceptual thresholds in these high end regions and plays an important role in perceptual bitrate ladder construction. However, the relationship between JND and the more widely used Mean Opinion Score (MOS) remains unclear. In this paper, we conduct a Degradation Category Rating (DCR) subjective study based on an existing JND dataset to examine how MOS corresponds to the 75% Satisfied User Ratio (SUR) points of the 1st and 2nd JNDs. We find that while MOS values at JND points generally align with theoretical expectations (e.g., 4.75 for the 75% SUR of the 1st JND), the reverse mapping from MOS to JND is ambiguous due to overlapping confidence intervals across PVS indices. Statistical significance analysis further shows that DCR studies with limited participants may not detect meaningful differences between reference and JND videos.

Is there a relationship between Mean Opinion Score (MOS) and Just Noticeable Difference (JND)?

TL;DR

This study addresses how Mean Opinion Score (MOS) relates to Just Noticeable Difference (JND) in high-quality video regions, a key concern for perceptual bitrate ladders. It uses a Degradation Category Rating (DCR) protocol on data derived from the HD-VJND JND dataset, modeling JND distributions with Satisfied User Ratio (SUR) and examining MOS at the 75% SUR anchors of the 1st and 2nd JND. The results show MOS values near ideal targets, with the 1st JND anchor at 1080p yielding about and the 2nd JND decreasing with increased distortion, e.g., at the ideal anchor; importantly, the mapping is one-way: MOS can inform the JND level but cannot uniquely determine it due to overlapping confidence intervals and limited study power. These findings have practical implications for perceptual bitrate ladder design and emphasize the need for larger-scale subjective studies to robustly connect MOS and JND in premium video experiences.

Abstract

Evaluating perceived video quality is essential for ensuring high Quality of Experience (QoE) in modern streaming applications. While existing subjective datasets and Video Quality Metrics (VQMs) cover a broad quality range, many practical use cases especially for premium users focus on high quality scenarios requiring finer granularity. Just Noticeable Difference (JND) has emerged as a key concept for modeling perceptual thresholds in these high end regions and plays an important role in perceptual bitrate ladder construction. However, the relationship between JND and the more widely used Mean Opinion Score (MOS) remains unclear. In this paper, we conduct a Degradation Category Rating (DCR) subjective study based on an existing JND dataset to examine how MOS corresponds to the 75% Satisfied User Ratio (SUR) points of the 1st and 2nd JNDs. We find that while MOS values at JND points generally align with theoretical expectations (e.g., 4.75 for the 75% SUR of the 1st JND), the reverse mapping from MOS to JND is ambiguous due to overlapping confidence intervals across PVS indices. Statistical significance analysis further shows that DCR studies with limited participants may not detect meaningful differences between reference and JND videos.
Paper Structure (10 sections, 4 equations, 4 figures, 4 tables)

This paper contains 10 sections, 4 equations, 4 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Subjective study design overview
  3. HD-VJND dataset
  4. Modeling JND Distributions with SUR
  5. DCR-Based MOS Collection
  6. Results and analysis
  7. Data screening
  8. MOS at JND Anchor Points
  9. Statistical Significance Analysis
  10. Conclusion

Figures (4)

  • Figure 1: The relationship between MOS and JND.
  • Figure 2: Illustration of the Satisfied User Ratio (SUR) for JND.
  • Figure 3: Bias and inconsistency of the 20 participants in the DCR dataset. Participants sharing the same home environment are indicated by the same color, while the same marker shape denotes the same TV reference.
  • Figure 4: DCR MOS value for 75%SUR in 1st JND and 2nd JND in 1080p and 2nd JND in 720p. (Note: the original MOS are mapped to make the reference MOS = 5)