Visual Attention Modeling for Stereoscopic Video: A Benchmark and Computational Model

doi:10.1109/TIP.2017.2721112

Paper

Visual Attention Modeling for Stereoscopic Video: A Benchmark and Computational Model

Published Jun 28, 2017 · Yuming Fang, Chi Zhang, Jing Li

IEEE Transactions on Image Processing

Q1 SJR score

53

Citations

3

Influential Citations

Full text

Semantic Scholar

Abstract

In this paper, we investigate the visual attention modeling for stereoscopic video from the following two aspects. First, we build one large-scale eye tracking database as the benchmark of visual attention modeling for stereoscopic video. The database includes 47 video sequences and their corresponding eye fixation data. Second, we propose a novel computational model of visual attention for stereoscopic video based on Gestalt theory. In the proposed model, we extract the low-level features, including luminance, color, texture, and depth, from discrete cosine transform coefficients, which are used to calculate feature contrast for the spatial saliency computation. The temporal saliency is calculated by the motion contrast from the planar and depth motion features in the stereoscopic video sequences. The final saliency is estimated by fusing the spatial and temporal saliency with uncertainty weighting, which is estimated by the laws of proximity, continuity, and common fate in Gestalt theory. Experimental results show that the proposed method outperforms the state-of-the-art stereoscopic video saliency detection models on our built large-scale eye tracking database and one other database (DML-ITRACK-3D).

Study Snapshot

Key takeawayThe proposed computational model for visual attention in stereoscopic video, based on Gestalt theory, outperforms state-of-the-art models in detecting saliency in large-scale eye tracking databases.

PopulationOlder adults (50-71 years)

Sample size24

MethodsObservational

OutcomesBody Mass Index projections

ResultsSocial networks mitigate obesity in older groups.

Sign up to use Study Snapshot

Consensus is limited without an account. Create an account or sign in to get more searches and use the Study Snapshot.

Create an account

Paper

Visual Attention Modeling for Stereoscopic Video: A Benchmark and Computational Model

Sign up to use Study Snapshot

References

Online object tracking based on CNN with spatial-temporal saliency guided sampling

This work improves object tracking performance by incorporating spatial-temporal saliency detection, guiding more accurate target localization and sampling within an inter-frame motion flow map.

Medical media analytics via ranking and big learning: A multi-modality image-based disease severity prediction study

The proposed computer-aided diagnosis strategy, combining ranking and learning techniques, effectively predicts dementia disease severity in multi-modality MRI images, outperforming existing methods.

A Universal Framework for Salient Object Detection

The proposed universal framework enhances the performance of existing saliency detection methods by using distance weighting, adaptive binarization, and morphological closing, and an iterative optimization method.

Saliency Detection for Stereoscopic Images Based on Depth Confidence Analysis and Multiple Cues Fusion

The proposed method improves saliency detection in stereoscopic images by incorporating depth information and multiple cues fusion, outperforming existing methods.

A proto-object based saliency model in three-dimensional space

Adding depth information to a proto-object based saliency model improves its ability to predict perceptual saliency (eye fixations) in natural scenes.

Citations