PointGrounding
multimodal
+
+
+
+
About
PointGrounding is a multimodal benchmark that evaluates AI models' ability to ground language descriptions to specific spatial locations in visual scenes through pointing tasks. This comprehensive evaluation tests visual-linguistic understanding, spatial reasoning, and precise localization capabilities, challenging models to accurately identify and point to objects or regions described in natural language within complex visual environments.
+
+
+
+
Evaluation Stats
Total Models1
Organizations1
Verified Results0
Self-Reported1
+
+
+
+
Benchmark Details
Max Score1
Language
en
+
+
+
+
Performance Overview
Score distribution and top performers
Score Distribution
1 models
Top Score
66.5%
Average Score
66.5%
High Performers (80%+)
0Top Organizations
#1Alibaba Cloud / Qwen Team
1 model
66.5%
+
+
+
+
Leaderboard
1 models ranked by performance on PointGrounding
| License | Links | ||||
|---|---|---|---|---|---|
| Mar 27, 2025 | Apache 2.0 | 66.5% |
