RefVOS: A Closer Look at Referring Expressions for Video Object Segmentation

TL;DR

This paper introduces RefVOS, a neural network for language-guided video object segmentation, highlighting the importance of non-trivial referring expressions and analyzing challenges related to motion and actions.

Contribution

The work provides a new categorization of referring expressions and demonstrates RefVOS's effectiveness, revealing key challenges in understanding motion and static actions.

Findings

RefVOS achieves state-of-the-art results in language-guided VOS.

Non-trivial REs are more challenging and reveal limitations in current models.

Understanding motion and static actions are major challenges for the task.

Abstract

The task of video object segmentation with referring expressions (language-guided VOS) is to, given a linguistic phrase and a video, generate binary masks for the object to which the phrase refers. Our work argues that existing benchmarks used for this task are mainly composed of trivial cases, in which referents can be identified with simple phrases. Our analysis relies on a new categorization of the phrases in the DAVIS-2017 and Actor-Action datasets into trivial and non-trivial REs, with the non-trivial REs annotated with seven RE semantic categories. We leverage this data to analyze the results of RefVOS, a novel neural network that obtains competitive results for the task of language-guided image segmentation and state of the art results for language-guided VOS. Our study indicates that the major challenges for the task are related to understanding motion and static actions.