AGIdex
agidex/Capabilities/Vision/Scene Understanding
SUB-CAPABILITY · VISION

Scene Understanding

76%
vs. HUMAN BASELINE = 100
PartialConf · High

Scoring rubric

0–20
Early research
20–40
Narrow benchmark competence
40–60
Strong benchmark · reliability gaps
60–80
Human-competitive in common scenarios
80–100
Comparable to typical skilled adult
100+
Reliably exceeds typical human baseline

Score vs. baseline

Trend · Last 12 months
12mo ago   %
6mo ago   %
Today   %
Δ 12mo   +0

What this measures

Human baseline

Describe what is happening in a photograph — objects, actions, relationships.

Human frontier

Infer intent, causality, and off-frame context from a single image.

Current state

What works

Dense image captioning and object relationship detection on standard datasets.

Key gaps

Causal scene inference; understanding what happened before and after a frame.

Evidence

0 sources
TechnologyQualitySourceScore vs. baselineScore
No evidence records yet.