# DS-Net: Dynamic Spatiotemporal Network for Video Salient Object Detection

arXiv: https://arxiv.org/abs/2012.04886 · HTML: https://mlsanity.com/paper/2012.04886/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 9 Dec 2020.

## Built on by (most cited first, 8 of 8)

- [Salient Object Detection via Integrity Learning](https://mlsanity.com/paper/2101.07663/) (2021)
- [PSNet: Parallel Symmetric Network for Video Salient Object Detection](https://mlsanity.com/paper/2210.05912/) (2022)
- [Salient Object Detection in RGB-D Videos](https://mlsanity.com/paper/2310.15482/) (2023)
- [Turb-Seg-Res: A Segment-then-Restore Pipeline for Dynamic Videos with Atmospheric Turbulence](https://mlsanity.com/paper/2404.13605/) (2024)
- [UniST: Towards Unifying Saliency Transformer for Video Saliency Prediction and Detection](https://mlsanity.com/paper/2309.08220/) (2023)
- [Efficient Unsupervised Video Object Segmentation Network Based on Motion Guidance](https://mlsanity.com/paper/2211.05364/) (2022)
- [A Spatial-Temporal Dual-Mode Mixed Flow Network for Panoramic Video Salient Object Detection](https://mlsanity.com/paper/2310.09016/) (2023)
- [Unsupervised Region-Growing Network for Object Segmentation in Atmospheric Turbulence](https://mlsanity.com/paper/2311.03572/) (2023)

## Abstract (arXiv)

As moving objects always draw more attention of human eyes, the temporal motive information is always exploited complementarily with spatial information to detect salient objects in videos. Although efficient tools such as optical flow have been proposed to extract temporal motive information, it often encounters difficulties when used for saliency detection due to the movement of camera or the partial movement of salient objects. In this paper, we investigate the complimentary roles of spatial and temporal information and propose a novel dynamic spatiotemporal network (DS-Net) for more effective fusion of spatiotemporal information. We construct a symmetric two-bypass network to explicitly extract spatial and temporal features. A dynamic weight generator (DWG) is designed to automatically learn the reliability of corresponding saliency branch. And a top-down cross attentive aggregation (CAA) procedure is designed so as to facilitate dynamic complementary aggregation of spatiotemporal features. Finally, the features are modified by spatial attention with the guidance of coarse saliency map and then go through decoder part for final saliency map. Experimental results on five benchmarks VOS, DAVIS, FBMS, SegTrack-v2, and ViSal demonstrate that the proposed method achieves superior performance than state-of-the-art algorithms. The source code is available at https://github.com/TJUMMG/DS-Net.