Rethinking Space-Time Networks with Improved Memory Coverage for Efficient Video Object Segmentation

TL;DR

This paper introduces a novel space-time correspondence method for video object segmentation that improves memory utilization and inference speed by using Euclidean distance for affinity, achieving state-of-the-art results.

Contribution

It proposes a new affinity measure and a diversified voting scheme that enhance memory efficiency and accuracy in video object segmentation.

Findings

Achieves state-of-the-art results on DAVIS and YouTubeVOS datasets.

Runs at over 20 FPS for multiple objects.

Improves memory utilization and inference robustness.

Abstract

This paper presents a simple yet effective approach to modeling space-time correspondences in the context of video object segmentation. Unlike most existing approaches, we establish correspondences directly between frames without re-encoding the mask features for every object, leading to a highly efficient and robust framework. With the correspondences, every node in the current query frame is inferred by aggregating features from the past in an associative fashion. We cast the aggregation process as a voting problem and find that the existing inner-product affinity leads to poor use of memory with a small (fixed) subset of memory nodes dominating the votes, regardless of the query. In light of this phenomenon, we propose using the negative squared Euclidean distance instead to compute the affinities. We validated that every memory node now has a chance to contribute, and experimentally showed that such diversified voting is beneficial to both memory efficiency and inference accuracy. The synergy of correspondence networks and diversified voting works exceedingly well, achieves new state-of-the-art results on both DAVIS and YouTubeVOS datasets while running significantly faster at 20+ FPS for multiple objects without bells and whistles.