Recall@50: 32.9Recall@100: 36.0mean Recall @100: 14.1R@100: 36.0
Papers with Code · CC BY-SA 4.0
Scene Parsing
Scene parsing is to segment and parse an image into different image regions associated with semantic categories, such as sky, road, person, and bed. MIT Description
논문
Top-5 Accuracy: 0.87
F1: 0.75
mIoU: 48.04%mAP: 68.96%
mean Recall @20: 11.63mean Recall @100: 17.24
mean Recall @20: 7.9
Median Relighting Error: 1.25
Recall@50: 31.74mean Recall @20: 7.1
R@100: 39.12
Recall@50: 18.32
Recall@50: 23.5Recall@100: 27.2mean Recall @100: 18.0
Recall@50: 28.9
Recall@50: 11.4
Recall@50: 27.1
LABRAD-OR: Lightweight Memory Scene Graphs for Accurate Bimodal Reasoning in Dynamic Operating Rooms
F1: 0.88
Recall@50: 27.9
F1: 0.90
F1: 0.901
NeuSyRE: Neuro-Symbolic Visual Understanding and Reasoning Framework based on Scene Graph Enrichment
R@100: 39.1mR@100: 12.6mR@50: 10.9
Recall@50: 27.7Recall@20: 21.6
F1: 0.91
Recall@50: 23.0mean Recall @20: 12.3mean Recall @100: 19.2mR@100: 19.2
mIoU: 47.87%mAP: 62.69%
Recall@50: 25.4
Recall@50: 10.72
Recall@50: 28.2Recall@20: 22.2Recall@100: 31.4
Recall@50: 31.93mean Recall @20: 6.9
Recall@50: 18.16