# EpO-Net: Exploiting Geometric Constraints on Dense Trajectories for Motion Saliency

arXiv: https://arxiv.org/abs/1909.13258 · HTML: https://mlsanity.com/paper/1909.13258/

## Key facts

- Cited by 22 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Sep 2019.

## Built on by (most cited first, 22 of 22)

- [Full-Duplex Strategy for Video Object Segmentation](https://mlsanity.com/paper/2108.03151/) (2021)
- [Hierarchical Feature Alignment Network for Unsupervised Video Object Segmentation](https://mlsanity.com/paper/2207.08485/) (2022)
- [Segment Any Motion in Videos](https://mlsanity.com/paper/2503.22268/) (2025)
- [Iteratively Selecting an Easy Reference Frame Makes Unsupervised Video Object Segmentation Easier](https://mlsanity.com/paper/2112.12402/) (2021)
- [Target-Aware Object Discovery and Association for Unsupervised Video Multi-Object Segmentation](https://mlsanity.com/paper/2104.04782/) (2021)
- [Multi-Source Fusion and Automatic Predictor Selection for Zero-Shot Video Object Segmentation](https://mlsanity.com/paper/2108.05076/) (2021)
- [Video Salient Object Detection via Contrastive Features and Attention Modules](https://mlsanity.com/paper/2111.02368/) (2021)
- [SimulFlow: Simultaneously Extracting Feature and Identifying Target for Unsupervised Video Object Segmentation](https://mlsanity.com/paper/2311.18286/) (2023)
- [Implicit Motion-Compensated Network for Unsupervised Video Object Segmentation](https://mlsanity.com/paper/2204.02791/) (2022)
- [Co-attention Propagation Network for Zero-Shot Video Object Segmentation](https://mlsanity.com/paper/2304.03910/) (2023)
- [Adaptive Multi-source Predictor for Zero-shot Video Object Segmentation](https://mlsanity.com/paper/2303.10383/) (2023)
- [Online Unsupervised Video Object Segmentation via Contrastive Motion Clustering](https://mlsanity.com/paper/2306.12048/) (2023)
- [Motion Segmentation from a Moving Monocular Camera](https://mlsanity.com/paper/2309.13772/) (2023)
- [Zero-Shot Monocular Motion Segmentation in the Wild by Combining Deep Learning with Geometric Motion Model Fusion](https://mlsanity.com/paper/2405.01723/) (2024)
- [Multiscale Memory Comparator Transformer for Few-Shot Video Segmentation](https://mlsanity.com/paper/2307.07812/) (2023)
- [Hierarchical Graph Pattern Understanding for Zero-Shot VOS](https://mlsanity.com/paper/2312.09525/) (2023)
- [Divided Attention: Unsupervised Multi-Object Discovery with Contextually Separated Slots](https://mlsanity.com/paper/2304.01430/) (2023)
- [MED-VT++: Unifying Multimodal Learning with a Multiscale Encoder-Decoder Video Transformer](https://mlsanity.com/paper/2304.05930/) (2023)
- [Video Decomposition Prior: A Methodology to Decompose Videos into Layers](https://mlsanity.com/paper/2412.04930/) (2024)
- [CAMotion: A High-Quality Benchmark for Camouflaged Moving Object Detection in the Wild](https://mlsanity.com/paper/2604.08287/) (2026)
- [Towards Imbalanced Motion: Part-Decoupling Network for Video Portrait Segmentation](https://mlsanity.com/paper/2307.16565/) (2023)
- [Dense Monocular Motion Segmentation Using Optical Flow and Pseudo Depth Map: A Zero-Shot Approach](https://mlsanity.com/paper/2406.18837/) (2024)

## Abstract (arXiv)

The existing approaches for salient motion segmentation are unable to explicitly learn geometric cues and often give false detections on prominent static objects. We exploit multiview geometric constraints to avoid such shortcomings. To handle the nonrigid background like a sea, we also propose a robust fusion mechanism between motion and appearance-based features. We find dense trajectories, covering every pixel in the video, and propose trajectory-based epipolar distances to distinguish between background and foreground regions. Trajectory epipolar distances are data-independent and can be readily computed given a few features' correspondences between the images. We show that by combining epipolar distances with optical flow, a powerful motion network can be learned. Enabling the network to leverage both of these features, we propose a simple mechanism, we call input-dropout. Comparing the motion-only networks, we outperform the previous state of the art on DAVIS-2016 dataset by 5.2% in the mean IoU score. By robustly fusing our motion network with an appearance network using the input-dropout mechanism, we also outperform the previous methods on DAVIS-2016, 2017 and Segtrackv2 dataset.