An Empirical Study of Propagation-based Methods for Video Object Segmentation

Abstract

While propagation-based approaches have achieved state-of-the-art performance for video object segmentation, the literature lacks a fair comparison of different methods using the same settings. In this paper, we carry out an empirical study for propagation-based methods. We view these approaches from a unified perspective and conduct detailed ablation study for core methods, input cues, multi-object combination and training strategies. With careful designs, our improved end-to-end memory networks achieve a global mean of 76.1 on DAVIS 2017 val set.

arXiv commentThe 2019 DAVIS Challenge on Video Object Segmentation - CVPR Workshops

References in corpus (5)

• YouTube-VOS: A Large-Scale Video Object Segmentation Benchmark
• Video Object Segmentation with Re-identification
• FEELVOS: Fast End-to-End Embedding Learning for Video Object Segmentation
• MHP-VOS: Multiple Hypotheses Propagation for Video Object Segmentation
• Discriminative Online Learning for Fast Video Object Segmentation

Cited by in corpus (1)

• Contextual Guided Segmentation Framework for Semi-supervised Video Instance Segmentation