Region Aware Video Object Segmentation with Deep Motion Modeling

TL;DR

This paper introduces RAVOS, a region-aware approach for video object segmentation that improves efficiency and accuracy by focusing on regions of interest and incorporating a new occlusion benchmark dataset.

Contribution

The paper proposes a novel region-aware VOS method with motion-based ROI prediction and motion path memory, along with a new large-scale occlusion dataset, OVOS.

Findings

Achieves state-of-the-art performance on DAVIS and YouTube-VOS benchmarks.

Runs at 42 FPS on DAVIS and 23 FPS on YouTube-VOS with high accuracy.

Outperforms existing methods in efficiency and occlusion handling.

Abstract

Current semi-supervised video object segmentation (VOS) methods usually leverage the entire features of one frame to predict object masks and update memory. This introduces significant redundant computations. To reduce redundancy, we present a Region Aware Video Object Segmentation (RAVOS) approach that predicts regions of interest (ROIs) for efficient object segmentation and memory storage. RAVOS includes a fast object motion tracker to predict their ROIs in the next frame. For efficient segmentation, object features are extracted according to the ROIs, and an object decoder is designed for object-level segmentation. For efficient memory storage, we propose motion path memory to filter out redundant context by memorizing the features within the motion path of objects between two frames. Besides RAVOS, we also propose a large-scale dataset, dubbed OVOS, to benchmark the performance of VOS models under occlusions. Evaluation on DAVIS and YouTube-VOS benchmarks and our new OVOS dataset show that our method achieves state-of-the-art performance with significantly faster inference time, e.g., 86.1 J&F at 42 FPS on DAVIS and 84.4 J&F at 23 FPS on YouTube-VOS.