Tackling Background Distraction in Video Object Segmentation

TL;DR

This paper introduces three novel strategies to improve semi-supervised video object segmentation by effectively suppressing background distractors, achieving state-of-the-art performance with real-time efficiency.

Contribution

The paper presents a new framework with diversified template construction, a learnable distance-scoring function, and swap-and-attach augmentation for better distractor suppression in VOS.

Findings

Achieves comparable performance to state-of-the-art methods.

Operates in real-time with high accuracy.

Demonstrates superiority through qualitative results.

Abstract

Semi-supervised video object segmentation (VOS) aims to densely track certain designated objects in videos. One of the main challenges in this task is the existence of background distractors that appear similar to the target objects. We propose three novel strategies to suppress such distractors: 1) a spatio-temporally diversified template construction scheme to obtain generalized properties of the target objects; 2) a learnable distance-scoring function to exclude spatially-distant distractors by exploiting the temporal consistency between two consecutive frames; 3) swap-and-attach augmentation to force each object to have unique features by providing training samples containing entangled objects. On all public benchmark datasets, our model achieves a comparable performance to contemporary state-of-the-art approaches, even with real-time performance. Qualitative results also demonstrate the superiority of our approach over existing methods. We believe our approach will be widely used for future VOS research.