Efficient Video Object Segmentation via Network Modulation

TL;DR

This paper introduces a fast, one-pass method for video object segmentation that adapts to specific objects using a meta neural network, significantly reducing processing time while maintaining high accuracy.

Contribution

A novel network modulation approach using a meta neural network enables rapid adaptation of segmentation models in a single forward pass, outperforming fine-tuning methods in speed.

Findings

70 times faster than fine-tuning methods

Achieves similar accuracy to state-of-the-art approaches

Effective with limited visual and spatial information

Abstract

Video object segmentation targets at segmenting a specific object throughout a video sequence, given only an annotated first frame. Recent deep learning based approaches find it effective by fine-tuning a general-purpose segmentation model on the annotated frame using hundreds of iterations of gradient descent. Despite the high accuracy these methods achieve, the fine-tuning process is inefficient and fail to meet the requirements of real world applications. We propose a novel approach that uses a single forward pass to adapt the segmentation model to the appearance of a specific object. Specifically, a second meta neural network named modulator is learned to manipulate the intermediate layers of the segmentation network given limited visual and spatial information of the target object. The experiments show that our approach is 70times faster than fine-tuning approaches while achieving similar accuracy.