Video Object Segmentation with Language Referring Expressions
This addresses the practical issue of reducing annotation costs for video object segmentation, making it more accessible for applications like video editing or surveillance, though it is incremental as it builds on existing language grounding models.
The paper tackles the problem of expensive pixel-accurate masks in video object segmentation by using language referring expressions to identify target objects, achieving performance on par with mask-based methods on DAVIS'16 and competitive with scribble-based methods on DAVIS'17.
Most state-of-the-art semi-supervised video object segmentation methods rely on a pixel-accurate mask of a target object provided for the first frame of a video. However, obtaining a detailed segmentation mask is expensive and time-consuming. In this work we explore an alternative way of identifying a target object, namely by employing language referring expressions. Besides being a more practical and natural way of pointing out a target object, using language specifications can help to avoid drift as well as make the system more robust to complex dynamics and appearance variations. Leveraging recent advances of language grounding models designed for images, we propose an approach to extend them to video data, ensuring temporally coherent predictions. To evaluate our method we augment the popular video object segmentation benchmarks, DAVIS'16 and DAVIS'17 with language descriptions of target objects. We show that our language-supervised approach performs on par with the methods which have access to a pixel-level mask of the target object on DAVIS'16 and is competitive to methods using scribbles on the challenging DAVIS'17 dataset.