Omnidirectional Image Quality Captioning: A Large-scale Database and A New Model
This work addresses the challenge of omnidirectional image quality assessment for applications like VR and 360-degree imaging, though it is incremental as it builds on existing OIQA methods by extending them to heterogeneous distortions.
The authors tackled the problem of assessing the quality of omnidirectional images with heterogeneous distortions by creating a large-scale database (OIQ-10K) of 10,000 images and proposing a new model (IQCaption360) that generates quality captions, which outperforms state-of-the-art methods by a significant margin.
The fast growing application of omnidirectional images calls for effective approaches for omnidirectional image quality assessment (OIQA). Existing OIQA methods have been developed and tested on homogeneously distorted omnidirectional images, but it is hard to transfer their success directly to the heterogeneously distorted omnidirectional images. In this paper, we conduct the largest study so far on OIQA, where we establish a large-scale database called OIQ-10K containing 10,000 omnidirectional images with both homogeneous and heterogeneous distortions. A comprehensive psychophysical study is elaborated to collect human opinions for each omnidirectional image, together with the spatial distributions (within local regions or globally) of distortions, and the head and eye movements of the subjects. Furthermore, we propose a novel multitask-derived adaptive feature-tailoring OIQA model named IQCaption360, which is capable of generating a quality caption for an omnidirectional image in a manner of textual template. Extensive experiments demonstrate the effectiveness of IQCaption360, which outperforms state-of-the-art methods by a significant margin on the proposed OIQ-10K database. The OIQ-10K database and the related source codes are available at https://github.com/WenJuing/IQCaption360.