Few-Shot Segmentation Propagation with Guided Networks

TL;DR

This paper introduces guided networks for few-shot segmentation that efficiently adapt to new tasks with minimal supervision, enabling real-time interactive and video segmentation with state-of-the-art accuracy.

Contribution

The paper presents a novel guided network architecture that can switch tasks and update quickly with minimal supervision, advancing few-shot segmentation capabilities.

Findings

Achieves state-of-the-art accuracy with minimal annotation

Enables real-time interactive video segmentation

Supports segmentation across space, time, and scenes

Abstract

Learning-based methods for visual segmentation have made progress on particular types of segmentation tasks, but are limited by the necessary supervision, the narrow definitions of fixed tasks, and the lack of control during inference for correcting errors. To remedy the rigidity and annotation burden of standard approaches, we address the problem of few-shot segmentation: given few image and few pixel supervision, segment any images accordingly. We propose guided networks, which extract a latent task representation from any amount of supervision, and optimize our architecture end-to-end for fast, accurate few-shot segmentation. Our method can switch tasks without further optimization and quickly update when given more guidance. We report the first results for segmentation from one pixel per concept and show real-time interactive video segmentation. Our unified approach propagates pixel annotations across space for interactive segmentation, across time for video segmentation, and across scenes for semantic segmentation. Our guided segmentor is state-of-the-art in accuracy for the amount of annotation and time. See http://github.com/shelhamer/revolver for code, models, and more details.