# Treating Motion as Option with Output Selection for Unsupervised Video Object Segmentation

arXiv: https://arxiv.org/abs/2309.14786 · HTML: https://mlsanity.com/paper/2309.14786/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 48 of its 86 references that are in the map.
- First posted on arXiv 26 Sep 2023.

## Builds on (most cited first, 30 of 48)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1802.02611/) (2018)
- [SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers](https://mlsanity.com/paper/2105.15203/) (2021)
- [CBAM: Convolutional Block Attention Module](https://mlsanity.com/paper/1807.06521/) (2018)
- [RAFT: Recurrent All-Pairs Field Transforms for Optical Flow](https://mlsanity.com/paper/2003.12039/) (2020)
- [Rethinking Atrous Convolution for Semantic Image Segmentation](https://mlsanity.com/paper/1706.05587/) (2017)
- [The Kinetics Human Action Video Dataset](https://mlsanity.com/paper/1705.06950/) (2017)
- [Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?](https://mlsanity.com/paper/1711.09577/) (2017)
- [Video Object Segmentation using Space-Time Memory Networks](https://mlsanity.com/paper/1904.00607/) (2019)
- [YouTube-VOS: A Large-Scale Video Object Segmentation Benchmark](https://mlsanity.com/paper/1809.03327/) (2018)
- [Video Classification with Channel-Separated Convolutional Networks](https://mlsanity.com/paper/1904.02811/) (2019)
- [The Secrets of Salient Object Segmentation](https://mlsanity.com/paper/1406.2807/) (2014)
- [Collaborative Video Object Segmentation by Foreground-Background Integration](https://mlsanity.com/paper/2003.08333/) (2020)
- [See More, Know More: Unsupervised Video Object Segmentation with Co-Attention Siamese Networks](https://mlsanity.com/paper/2001.06810/) (2020)
- [Kernelized Memory Network for Video Object Segmentation](https://mlsanity.com/paper/2007.08270/) (2020)
- [Video Object Segmentation with Adaptive Feature Bank and Uncertain-Region Refinement](https://mlsanity.com/paper/2010.07958/) (2020)
- [RVOS: End-to-End Recurrent Network for Video Object Segmentation](https://mlsanity.com/paper/1903.05612/) (2019)
- [Hierarchically Decomposed Graph Convolutional Networks for Skeleton-Based Action Recognition](https://mlsanity.com/paper/2208.10741/) (2022)
- [Zero-Shot Video Object Segmentation via Attentive Graph Neural Networks](https://mlsanity.com/paper/2001.06807/) (2020)
- [Full-Duplex Strategy for Video Object Segmentation](https://mlsanity.com/paper/2108.03151/) (2021)
- [A Generative Appearance Model for End-to-end Video Object Segmentation](https://mlsanity.com/paper/1811.11611/) (2018)
- [Making a Case for 3D Convolutions for Object Segmentation in Videos](https://mlsanity.com/paper/2008.11516/) (2020)
- [Anchor Diffusion for Unsupervised Video Object Segmentation](https://mlsanity.com/paper/1910.10895/) (2019)
- [Hierarchical Feature Alignment Network for Unsupervised Video Object Segmentation](https://mlsanity.com/paper/2207.08485/) (2022)
- [Learning Discriminative Feature with CRF for Unsupervised Video Object Segmentation](https://mlsanity.com/paper/2008.01270/) (2020)
- [Reliable Propagation-Correction Modulation for Video Object Segmentation](https://mlsanity.com/paper/2112.02853/) (2021)

## Abstract (arXiv)

Unsupervised video object segmentation aims to detect the most salient object in a video without any external guidance regarding the object. Salient objects often exhibit distinctive movements compared to the background, and recent methods leverage this by combining motion cues from optical flow maps with appearance cues from RGB images. However, because optical flow maps are often closely correlated with segmentation masks, networks can become overly dependent on motion cues during training, leading to vulnerability when faced with confusing motion cues and resulting in unstable predictions. To address this challenge, we propose a novel motion-as-option network that treats motion cues as an optional component rather than a necessity. During training, we randomly input RGB images into the motion encoder instead of optical flow maps, which implicitly reduces the network's reliance on motion cues. This design ensures that the motion encoder is capable of processing both RGB images and optical flow maps, leading to two distinct predictions depending on the type of input provided. To make the most of this flexibility, we introduce an adaptive output selection algorithm that determines the optimal prediction during testing.