Technical Standard
CHAPARRITO NALGÓN FRAMEWORK TECHNICAL STANDARD v1.1
Metrics, schemas, compliance criteria, and API rules for CNF-compliant fit-geometry evaluation.
1. Scope
Defines the rules, metrics, and data structures required for CNF-compliant fit-geometry evaluation. Applies to all scans, automated pipelines, and analyst tools producing fit-grade artifacts.
2. Terminology
- Contour Stability — silhouette smoothness and left/right symmetry score.
- Projection Vector — distance from pelvic reference plane to posterior surface along the normal axis.
- Curvature Radius — cross-sectional radius sampled at vertical levels on the seat contour.
- Tension Zones — classified cloth-body distance fields mapped to upper/mid/lower regions.
- FitBand — categorical fit output: tight / regular / relaxed / Levi’s-compatible.
- Mesh-Ready — flag indicating reconstruction accuracy and artifact cleanliness.
3. Input Requirements
- Input type: single RGB image.
- Minimum resolution:
720p. - Coverage: waist through mid-thigh, rear or lateral view.
- Acceptable poses: standing, seated, neutral stance.
- Metadata: submission ID, consent status, timestamp, analyst ID.
4. CV Processing Requirements
- 4.1 Segmentation: must isolate person, jeans, seat region, and thighs.
- 4.2 Pose Estimation: 2D keypoints → 3D skeleton via single-view lifting.
- 4.3 Body Mesh: SMPL-X fitting with silhouette alignment.
- 4.4 Cloth Mesh: canonical jeans mesh warped to body mesh using offset fields.
- 4.5 Depth/Normal Refinement: optional but recommended for curvature accuracy.
5. Metric Definitions
| Metric | Definition |
|---|---|
| Seat Curvature Radius | Cross-section curvature sampled at 3–5 vertical levels. |
| Projection Vector | Distance from pelvic reference plane to posterior mesh along surface normal. |
| Contour Stability | Silhouette smoothness plus left/right symmetry score. |
| Tension Readability | Wrinkle density + highlight intensity + seam displacement. |
| FitBand Classification | Cloth-body distance fields mapped to tight / regular / relaxed / Levi’s-compatible. |
6. Output Requirements
All CNF outputs must include the following artifacts:
cnffitgrade_report— seat grade, tier, score breakdown.cnfmeshspec— mesh validity, vertex count, reconstruction accuracy.cnftensionheatmap— tension zone intensities and classification.cnfcertlabel— final label with pass/fail flags and seal ID.
7. Data Storage Requirements
- Meshes: object storage with content-addressed keys.
- Metrics: relational DB with scan-level provenance records.
- Retention: retain raw images, intermediate masks, and final artifacts for audit.
8. Compliance
A system, scan, or model is CNF-compliant only if:
- All required metrics are computed and present.
- All output JSON schemas validate successfully.
- Reconstruction meets minimum silhouette accuracy thresholds for the requested tier.
- Provenance metadata is retained for audit and traceability.