Quality Index for Stereoscopic Images by Separately Evaluating Adding and Subtracting

Abstract

The human visual system (HVS) plays an important role in stereo image quality perception. Therefore, it has aroused many people's interest in how to take advantage of the knowledge of the visual perception in image quality assessment models. This paper proposes a full-reference metric for quality assessment of stereoscopic images based on the binocular difference channel and binocular summation channel. For a stereo pair, the binocular summation map and binocular difference map are computed first by adding and subtracting the left image and right image. Then the binocular summation is decoupled into two parts, namely additive impairments and detail losses. The quality of binocular summation is obtained as the adaptive combination of the quality of detail losses and additive impairments. The quality of binocular summation is computed by using the Contrast Sensitivity Function (CSF) and weighted multi-scale (MS-SSIM). Finally, the quality of binocular summation and binocular difference is integrated into an overall quality index. The experimental results indicate that compared with existing metrics, the proposed metric is highly consistent with the subjective quality assessment and is a robust measure. The result have also indirectly proved hypothesis of the existence of binocular summation and binocular difference channels.

Fig 1

Two alternative hypothesis of the process of stereopsis, (a) Schematic diagram of traditional view, (b) Schematic diagram of recent discovery.

Fig 2. Structure of the proposed metric.

Fig 3

Summation and difference of stereo pair (a) Left view, (b) Right view, (c) binocular difference map and (d) binocular summation map.

Fig 4. Subbands indexing of DWT.

Each subband is indexed by a level λ and an orientation θ. θ = 2, 3, 4 denote the vertical, horizontal, and diagonal subbands, respectively.

Fig 5. Decoupling Result of Test Binocular Summation.

Fig 6. CSF curve.

Fig 7. Weighting matrix w used in contrast masking.

Fig 8. 20 left views of original stereo image pairs.

Fig 9. One of the original left view images and the corresponding five types of distortion.

(a) Original left view. (b) JEPG compressed. (c) JEPG 2000 compressed. (d) White noise. (e) Blur. (f) Fast fading.

Fig 10

Scatter plots of objective scores versus subjective scores for the eight metric, (a)SSIM, (b)PSNR, (c)ADM, (d) Benoit’s scheme, (e) Chen’s scheme, and (f) Proposed scheme.

Fig 11. 10 statistic results of 18 distorted pairs of an original pair.

Fig 12. Performance comparison of each part and overall.