Video Sparse Transformer With Attention-Guided Memory for Video Object Detection

Open paper

Benchmark	Model	Rank	Results
object-detection-on-ua-detrac	VSTAM	#1	mAP: 90.39
video-instance-segmentation-on-youtube-vis-1	VSTAM	#24	mask AP: 39.0
video-object-detection-on-imagenet-vid	VSTAM	#4	MAP: 91.1