Sharper images.
Verified faces.
CivicSet builds two computer-vision products, Multi-Frame Super-Resolution and Facial Verification, delivered in-process with explainable, account-scoped results.
Multi-Frame Super-Resolution
Align, fuse, upscale
Facial Verification
Procrustes alignment
1:1 Only
No identification
In-Process Compute
No external plane
Two focused products for image processing
We build Multi-Frame Super-Resolution and Facial Verification, focused computer-vision products that return explainable, account-scoped results.
Multi-Frame Super-Resolution
Fuse a burst of low-quality frames of the same scene into one crisp, high-resolution image with sub-pixel alignment and robust fusion.
- Sub-pixel frame alignment
- Robust shift-and-add fusion
- Noise suppression through redundancy
- Alignment-error reporting
Facial Verification
Compare two faces for a same-or-different verdict using Procrustes geometric alignment. Verification only, never identification.
- Procrustes geometric alignment
- Embedding comparison
- Configurable match threshold
- Similarity and distance metrics
In-Process Compute
Both products run as server-only functions in the web runtime, with no container plane to deploy and no separate upload step.
- Server-side TypeScript engines
- Typed request contracts
- Size and pixel caps
- No external tool plane
Account-Scoped Storage
Private media is stored under account-scoped keys in Supabase Storage, governed by row-level security and account tenancy.
- Account-scoped object keys
- Row-level security
- Encrypted in transit and at rest
- Personal and team accounts
How it works
Choose a product
Pick Multi-Frame Super-Resolution to sharpen a burst of frames, or Facial Verification to compare two faces.
Provide inputs
Upload a burst of frames or a reference-and-probe pair. Inputs are validated against typed contracts before compute runs.
Review results
Inspect explainable outputs, alignment error, similarity, and distance, with your media stored account-scoped.
What guides us
Explainable by Design
Every result carries its metrics, alignment error, similarity, and distance, so outputs can be reviewed rather than trusted blindly.
Focused, Not Sprawling
Two products, done well. We concentrate on super-resolution and facial verification rather than a long tail of half-finished tools.
In-Process by Default
Compute runs inside the web runtime. There is no external tool plane, no separate deployment, and no hidden service behind the result.
Privacy-Respecting
Facial Verification is 1:1 only. There is no database matching surface and no identification of people from imagery.
2
Products
Multi-Frame Super-Resolution and Facial Verification
1:1
Verification
Face comparison only, no database identification
16
Max Frames
Frames fused per super-resolution run
In-process
Compute
Server-side, with no external tool plane
“Every run returns alignment error, similarity, and distance: metrics you can audit, not a black-box verdict.”
Explainable by design
Structured outputs on every run
Common questions
CivicSet builds two computer-vision products: Multi-Frame Super-Resolution, which fuses a burst of frames into one high-resolution image, and Facial Verification, which compares two faces for a same-or-different verdict.
Two products,
in the works
Multi-Frame Super-Resolution and Facial Verification are in active development. Leave your email and be first to know when they launch.