Tile Classification Based Viewport Prediction with Multi-modal Fusion Transformer

TL;DR

This paper introduces MFTR, a transformer-based method for viewport prediction in 360 videos that improves robustness and interpretability by classifying tiles as user interested or not, leveraging multi-modal data.

Contribution

The paper proposes a novel multi-modal fusion transformer approach for tile classification in viewport prediction, enhancing robustness and interpretability over trajectory-based methods.

Findings

MFTR outperforms state-of-the-art methods in accuracy and overlap ratio.

MFTR demonstrates better robustness and interpretability.

The method achieves competitive computational efficiency.

Abstract

Viewport prediction is a crucial aspect of tile-based 360 video streaming system. However, existing trajectory based methods lack of robustness, also oversimplify the process of information construction and fusion between different modality inputs, leading to the error accumulation problem. In this paper, we propose a tile classification based viewport prediction method with Multi-modal Fusion Transformer, namely MFTR. Specifically, MFTR utilizes transformer-based networks to extract the long-range dependencies within each modality, then mine intra- and inter-modality relations to capture the combined impact of user historical inputs and video contents on future viewport selection. In addition, MFTR categorizes future tiles into two categories: user interested or not, and selects future viewport as the region that contains most user interested tiles. Comparing with predicting head trajectories, choosing future viewport based on tile's binary classification results exhibits better robustness and interpretability. To evaluate our proposed MFTR, we conduct extensive experiments on two widely used PVS-HM and Xu-Gaze dataset. MFTR shows superior performance over state-of-the-art methods in terms of average prediction accuracy and overlap ratio, also presents competitive computation efficiency.