A self-attention model for viewport prediction based on distance constraint

CD Lan, X Qiu, C Miao, MT Zheng - The Visual Computer, 2023 - Springer
… existing viewport trajectory dataset, we find the randomness and continuity of the viewport
, we design a viewport prediction model based on a self-attention mechanism to provide more …

[PDF][PDF] Transformer-based Long-Term Viewport Prediction in 360° Video: Scanpath is All You Need.

FY Chao, C Ozcinar, A Smolic - MMSP, 2021 - people.rennes.inria.fr
… a novel viewport prediction model to address viewport prediction as a time … viewport
prediction model, called VPT360, which adopts the self-attention layer from transformer to predict

STAR-VP: Improving Long-term Viewport Prediction in 360° Videos via Space-aligned and Time-varying Fusion

B Gao, D Sheng, L Zhang, Q Qi, B He, Z Zhuang… - ACM Multimedia … - openreview.net
… The Encoder and Decoder have the same architecture but are based on self-attention and
cross-attention, respectively. C𝑓 (·) and C𝑡 (·) denote concatenating vectors along the feature …

Tile classification based viewport prediction with multi-modal fusion transformer

Z Zhang, Y Chen, W Zhang, C Yan, Q Zheng… - Proceedings of the 31st …, 2023 - dl.acm.org
Viewport prediction is a crucial aspect of tile-based 360 … a tile classification based viewport
prediction method with Multi-… and video contents on future viewport selection. In addition, …

Your attention is unique: Detecting 360-degree video saliency in head-mounted display for head movement prediction

A Nguyen, Z Yan, K Nahrstedt - Proceedings of the 26th ACM …, 2018 - dl.acm.org
… To fill in this gap, this paper shifts the traditional single-viewport saliency models that have
… using the new self-built dataset. Then the proposed head movement prediction framework is …

Machine learning based content-agnostic viewport prediction for 360-degree video

S Van Damme, MT Vega, F De Turck - ACM Transactions on Multimedia …, 2022 - dl.acm.org
… Therefore, the past years have seen growing attention to models for viewport prediction in
360֯ … Their method is evaluated on a self-created dataset for PHs of 1, 3 and 5 seconds. Their …

MADRL-Based Rate Adaptation for 360° Video Streaming With Multi-Viewpoint Prediction

H Wang, Z Long, H Dong… - IEEE Internet of Things …, 2024 - ieeexplore.ieee.org
viewport prediction to reduce bandwidth consumption. However, the performance of models
for single-viewpoint prediction is … [15] adopt LSTM and self-attention mechanism to predict

Synergistic Temporal-Spatial User-Aware Viewport Prediction for Optimal Adaptive 360-Degree Video Streaming

Y Wang, J Li, Z Li, S Shang, Y Liu - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
… and allocate their attention across the viewing area. The synergistic temporal-spatial …
This also proves the self-regulation ability of the proposed method. In the aforementioned …

MADRL-Based Rate Adaptation for 360$\degree $ Video Streaming with Multi-Viewpoint Prediction

H Wang, Z Long, H Dong, AE Saddik - arXiv preprint arXiv:2405.07759, 2024 - arxiv.org
viewport prediction to reduce bandwidth consumption. However, the performance of models
for single-viewpoint prediction is … [15] adopt LSTM and self-attention mechanism to predict

Toward visual behavior and attention understanding for augmented 360 degree videos

Y Zhu, X Min, D Zhu, G Zhai, X Yang, W Zhang… - ACM Transactions on …, 2023 - dl.acm.org
… To bridge the gap between the saliency prediction on real-world scenes and on scenes …
saliency prediction problem in AR. Local block images are extracted to simulate the viewport and …