Instance Embedding Transfer to Unsupervised Video Object Segmentation
This work addresses video object segmentation for computer vision applications, but it is incremental as it adapts existing image-based methods to videos without retraining.
The authors tackled unsupervised video object segmentation by transferring instance embeddings from static images to videos, achieving state-of-the-art performance on the DAVIS and FBMS datasets.
We propose a method for unsupervised video object segmentation by transferring the knowledge encapsulated in image-based instance embedding networks. The instance embedding network produces an embedding vector for each pixel that enables identifying all pixels belonging to the same object. Though trained on static images, the instance embeddings are stable over consecutive video frames, which allows us to link objects together over time. Thus, we adapt the instance networks trained on static images to video object segmentation and incorporate the embeddings with objectness and optical flow features, without model retraining or online fine-tuning. The proposed method outperforms state-of-the-art unsupervised segmentation methods in the DAVIS dataset and the FBMS dataset.