Evidence · first-party tested/Best AI Tools for Consistent AI Characters Across Scenes and Poses

The rooftop output preserves the short dark curly hair well.

✓ Worked🧾 artifact-verifiedinput + output shownTest date not recordedImagineArt →

What was measured

Accessory & detail retention

Retention of specific features — bindi, hair texture, skin tone, jewellery.

decisive for this rankingtransformation

Small identifying details like jewellery, bindi, hair texture, and skin tone are part of maintaining a consistent character identity across generations. (3 of 3 judges)

What was given, what came back

Test input: Near-profile occluded portrait stress test · image

Input — what we sent

Input not captured

This run recorded no prompt or input file for the test, so we cannot show you what produced the result below. Capture gaps are tracked, not hidden.

Near-profile portrait reference used as the maximum-difficulty stress test: face turned roughly 80–90 degrees, one eye partially occluded by fringe, short dark wavy hair, and natural outdoor lighting. Designed to probe whether tools can preserve identity under severe pose and occlusion.

Why this input is hard

· Near-profile identity preservation
· Occlusion handling
· Extreme pose consistency
· Lighting and hair-detail retention under stress

Output — unretouched

Also checked on this input — same tool, 2 other criteria

Identity preservation⚠ StruggledThe rooftop output does not maintain the near-profile stress condition, because the face is rendered more frontally visible than specified.Scene compliance⚠ StruggledThe rooftop output only partially follows the prompt: the sleeve length is wrong, the arms are asymmetrically raised, and the golden-hour lighting is only partly captured.

Provenance

Observation: 41976ac6-36c4-4195-be11-22377c2510f6
Evidence run: 1dfb8fa4-f7f0-47dc-911b-7db3af467e9a
Study: Generate Consistent AI Characters Across Different Scenes and Poses
Research task: 86b96df11
Tested at: not recorded
Source: first-party
Evidence state: verified
Proof shown: input + output shown
Cost / latency: not captured
Repeat run: not captured
Tester: not captured

The last three rows are honest blanks, not placeholders — our capture has no field for them yet.

Query this

get_evidence({
  tool: "imagineart"
})

MCP · mcp.aidemos.com/api/mcp

JSON · ai.aidemos.com/v1/observations/41976ac6-36c4-4195-be11-22377c2510f6

Free with attribution.

Same input, same check — 2 other tools

measured on Accessory & detail retention

ChatGPT✓ WorkedPreserves the short dark wavy hair’s volume and texture in the stress-test output, even though the face is lightly beautified.Gemini⚠ StruggledHair detail weakens under stress, with the output hair flatter and less wavy than the reference.

This evidence is published in

Best AI Tools for Consistent AI Characters Across Scenes and Poses →

Real inputs and real outputs, no retouching · every cell queryable via API & MCP · aidemos.com