Learning Cross-Modal Affinity for Referring Video Object Segmentation Targeting Limited Samples

TL;DR

This paper introduces a novel cross-modal affinity Transformer model for few-shot referring video object segmentation, enabling effective adaptation to new scenes with minimal annotations and establishing a new benchmark for the task.

Contribution

The paper proposes a cross-modal affinity module within a Transformer architecture for few-shot RVOS and creates a new benchmark to evaluate such methods.

Findings

Achieves state-of-the-art performance on the new FS-RVOS benchmark.

Outperforms baselines by 10% on Mini-Ref-YouTube-VOS.

Significantly better results on Mini-Ref-SAIL-VOS.

Abstract

Referring video object segmentation (RVOS), as a supervised learning task, relies on sufficient annotated data for a given scene. However, in more realistic scenarios, only minimal annotations are available for a new scene, which poses significant challenges to existing RVOS methods. With this in mind, we propose a simple yet effective model with a newly designed cross-modal affinity (CMA) module based on a Transformer architecture. The CMA module builds multimodal affinity with a few samples, thus quickly learning new semantic information, and enabling the model to adapt to different scenarios. Since the proposed method targets limited samples for new scenes, we generalize the problem as - few-shot referring video object segmentation (FS-RVOS). To foster research in this direction, we build up a new FS-RVOS benchmark based on currently available datasets. The benchmark covers a wide range and includes multiple situations, which can maximally simulate real-world scenarios. Extensive experiments show that our model adapts well to different scenarios with only a few samples, reaching state-of-the-art performance on the benchmark. On Mini-Ref-YouTube-VOS, our model achieves an average performance of 53.1 J and 54.8 F, which are 10% better than the baselines. Furthermore, we show impressive results of 77.7 J and 74.8 F on Mini-Ref-SAIL-VOS, which are significantly better than the baselines. Code is publicly available at https://github.com/hengliusky/Few_shot_RVOS.