Visual Scan-Path based Data-Augmentation for CNN-based 360-degree Image Quality Assessment

Abstract

360-degree Image quality assessment (IQA) is facing the major challenge of lack of ground-truth databases. This problem is accentuated for deep learning based approaches where the performances are as good as the available data. In this context, only two databases are used to train and validate deep learning-based IQA models. To compensate this lack, a dataaugmentation technique is investigated in this paper. We use visual scan-path to increase the learning examples from existing training data. Multiple scan-paths are predicted to account for the diversity of human observers. These scan-paths are then used to select viewports from the spherical representation. The results of the data-augmentation training scheme showed an improvement over not using it. We also try to answer the question of using the MOS obtained for the 360-degree image as the quality anchor for the whole set of extracted viewports in comparison to 2D blind quality metrics. The comparison showed the superiority of using the MOS when adopting a patch-based learning.

Patch-based CNN model for 360 image quality assessment with adaptive pooling strategies

“…Based on these observations, the parameter for both methods should be carefully chosen, as it is dependent on the variability and span of local qualities. In addition, the difference among OIQA and CVIQ is due to the nature and diversity of their content, as shown in [30]. This is also depicted by the provided curves, where an important gap between PLCC and SRCC values can be observed on CVIQ compared to OIQA independently of the used pooling methods.…”

“…By taking the content surrounding these fixations, we extract patches of 256 × 256 pixels. Selected patches are then extracted on the sphere in order to avoid any geometric distortions due to the sphere-to-plane projection [30,11]. In total, eighty patches are extracted from the 360-degree image I.…”