Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Grounded Text-to-Image Generation on LVIS (val)
Loading...
25.2
AP
Upper-bound
1.072
7.336
13.6
19.864
Jan 17, 2023
AP
AP (Rare)
AP (Common)
AP (Frequent)
Updated 3d ago
Evaluation Results
Method
Method
Links
AP
AP (Rare)
AP (Common)
AP (Frequent)
Upper-bound
2023.01
25.2
19
22.2
31.2
GLIGEN-LDM
Training data=GoldG,O3...
2023.01
11.1
9
9.8
13.4
GLIGEN-Stable
Training data=GoldG,O3...
2023.01
10.8
8.8
9.9
12.6
GLIGEN-LDM
Training data=GoldG,O365
2023.01
10.6
5.8
9.6
13.8
GLIGEN-LDM
Training data=COCO2014CD
2023.01
6.4
5.8
5.8
7.4
GLIGEN-LDM
Training data=COCO2014G
2023.01
6
4.4
6.1
6.6
GLIGEN-LDM
Training data=COCO2014D
2023.01
4.4
2.3
3.3
6.5
LAMA
Training data=LVIS
2023.01
2
0.9
1.3
3.2
Feedback
Search any
task
Search any
task