masktrack_rcnn

Video Instance Segmentation

Abstract

In this paper we present a new computer vision task, named video instance segmentation. The goal of this new task is simultaneous detection, segmentation and tracking of instances in videos. In words, it is the first time that the image instance segmentation problem is extended to the video domain. To facilitate research on this new task, we propose a large-scale benchmark called YouTube-VIS, which consists of 2883 high-resolution YouTube videos, a 40-category label set and 131k high-quality instance masks. In addition, we propose a novel algorithm called MaskTrack R-CNN for this task. Our new method introduces a new tracking branch to Mask R-CNN to jointly perform the detection, segmentation and tracking tasks simultaneously. Finally, we evaluate the proposed method and several strong baselines on our new dataset. Experimental results clearly demonstrate the advantages of the proposed algorithm and reveal insight for future improvement. We believe the video instance segmentation task will motivate the community along the line of research for video understanding.

Citation

@inproceedings{yang2019video,
  title={Video instance segmentation},
  author={Yang, Linjie and Fan, Yuchen and Xu, Ning},
  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
  pages={5188--5197},
  year={2019}
}

Results and models of MaskTrack R-CNN on YouTube-VIS 2019 validation dataset

As mentioned in Issues #6 in MaskTrack R-CNN, the result is kind of unstable for different trials, which ranges from 28 AP to 31 AP when using R-50-FPN as backbone. The checkpoint provided below is the best one from two experiments.

Method	Base detector	Backbone	Style	Lr schd	Mem (GB)	Inf time (fps)	AP	Config	Download
MaskTrack R-CNN	Mask R-CNN	R-50-FPN	pytorch	12e	1.61	-	30.2	config	model | log
MaskTrack R-CNN	Mask R-CNN	R-101-FPN	pytorch	12e	2.27	-	32.2	config	model | log
MaskTrack R-CNN	Mask R-CNN	X-101-FPN	pytorch	12e	3.69	-	34.7	config	model | log

Results and models of MaskTrack R-CNN on YouTube-VIS 2021 validation dataset

The checkpoint provided below is the best one from two experiments.

Method	Base detector	Backbone	Style	Lr schd	Mem (GB)	Inf time (fps)	AP	Config	Download
MaskTrack R-CNN	Mask R-CNN	R-50-FPN	pytorch	12e	1.61	-	28.7	config	model | log
MaskTrack R-CNN	Mask R-CNN	R-101-FPN	pytorch	12e	2.27	-	31.3	config	model | log
MaskTrack R-CNN	Mask R-CNN	X-101-FPN	pytorch	12e	3.69	-	33.5	config	model | log