Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
3D Object Captioning on Objaverse-LVIS 1k sampled
Loading...
88.7
CLIPScore
Tri-MARF
61.868
68.834
75.8
82.766
Jan 7, 2026
CLIPScore
A/B Score
ViLT R@5 (I2T)
ViLT R@5 (T2I)
Updated 4d ago
Evaluation Results
Method
Method
Links
CLIPScore
A/B Score
ViLT R@5 (I2T)
ViLT R@5 (T2I)
Tri-MARF
Speed (objects/hour)=12k
2026.01
88.7
-
45.2
43.8
Human Annotation
Speed (objects/hour)=0...
2026.01
82.4
2.3
40
38.5
ScoreAgg
Speed (objects/hour)=9k
2026.01
80.1
3.9
37.8
36
Cap3D
Speed (objects/hour)=8k
2026.01
78.6
3.3
35.2
33.4
3D-LLM
Speed (objects/hour)=6.5k
2026.01
77.4
3.2
34.9
33.3
ULIP-2
Speed (objects/hour)=7k
2026.01
75.2
3
33.1
31.5
PointCLIP
Speed (objects/hour)=5k
2026.01
65.3
2
22.4
20.8
Metadata
2026.01
65.2
1.5
20.1
18.7
GPT4Point
Speed (objects/hour)=4k
2026.01
62.9
1.8
18.7
17.1
Feedback
Search any
task
Search any
task