The cutout shows a visible cyan-blue halo around the backlit head and shoulders, so the edge is not a clean silhouette.
What was measured
Edge quality
How clean are the cutouts around hair, fingers, and clothing edges; are there visible halo, fringing, or jagged edges?
decisive for this rankingtransformation
Clean edges around hair, fingers, and clothing are a core measure of whether the background cutout works well at all. (3 of 3 judges)
What was given, what came back
Test input: Oregon Coast Walking · video · group: remove-or-replace-video-backgrounds-using-ai
Input — what we sent
Oregon Coast Walking
A daylight outdoor video of a single person walking along an Oregon coast beach, with the camera following the subject and ocean waves, sand, and sky visible in the background. It was used to test background removal and replacement on a moving subject in a natural outdoor setting.
Why this input is hard
- · Outdoor background removal
- · Subject segmentation during movement
- · Hair and edge preservation
- · AI background replacement quality
Output — unretouched

Also checked on this input — same tool, 2 other criteria
Lighting adaptation⚠ StruggledThe replacement background does not integrate lighting naturally: the subject’s rim light points from behind/above while the scene’s sun sits upper-left near the horizon, creating a direction mismatch.Output resolution✗ FailedIt downscales a 2160×3840 source to a 720×1280 export instead of preserving the original resolution.
Provenance
- Observation
- 74bae4c9-d90a-46bb-bd37-1e54aeddf263
- Evidence run
- 10155228-7ce8-438a-a210-547331ef080b
- Study
- Remove or Replace Video Backgrounds Using AI
- Research task
- 86ba42c2d
- 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: "descript",
scenario: "remove-or-replace-video-backgrounds-using-ai"
})MCP · mcp.aidemos.com/api/mcp
Free with attribution.
Same input, same check — 9 other tools
measured on Edge quality
Bria.ai⚠ StruggledA strong blue-white halo bleeds past the head and shoulders into the black background, so the silhouette is not clean at the edge.Cutout.Pro⚠ StruggledIntroduces a faint blue-teal halo around the head and shoulders on the cutout.FlexClip⚠ StruggledUnder harsh backlighting, the beach subject’s torso can read semi-transparent instead of fully opaque, so the matte is visibly incomplete beyond the recurring halo artifact.Fotor✓ WorkedThe moving subject is cut out cleanly across the walk, with the person remaining isolated rather than breaking into jagged edge artifacts.InVideo AI✓ WorkedIt kept the moving subject’s silhouette clean, with no visible green-screen fringing or halo around the edge.Kapwing⚠ StruggledOn the beach silhouette, the cutout edge is soft and fuzzy with a faint greenish fringe around the hood rather than a crisp line.Media.io⚠ StruggledNegative-space gaps between limbs and torso retained a faint warm brown tint instead of pure black, showing residual contamination rather than a clean cutout.Picsart⚠ StruggledOn the beach clip, backlit sun glare was not cleanly separated and remained as a rainbow-tinted haze over the head and shoulder instead of a clean silhouette.VEED⚠ StruggledLeaves visible haloing and softer torso edges on the cutout, so the silhouette does not resolve cleanly at the boundary.
This evidence is published in
Real inputs and real outputs, no retouching · every cell queryable via API & MCP · aidemos.com