Results Templates
Resource feasibility
| Frames | Resolution | Runs completed | Peak memory | Inference p50/p95 | Decision |
|---|---|---|---|---|---|
| 4 | 336 | ||||
| 3 | 280 |
Record the fallback row only when the primary fails.
Camera calibration
| Images accepted | Resolution | Mean reprojection error | Maximum error | Focus/exposure locked |
|---|---|---|---|---|
Single-window outputs
| Window | Valid depth | Confidence p40 | Retained points | Predicted fx/fy | PLY reload |
|---|---|---|---|---|---|
Window alignment
| Attempts | Accepted | Rejected | Success rate | Median correspondences | Median inlier ratio | Median residual |
|---|---|---|---|---|---|---|
For rejected attempts, include a separate count by rejection reason.
Doorway after scale alignment
| Dimension | Reference | Reconstructed | Absolute error | Percentage error |
|---|---|---|---|---|
| Width | ||||
| Height |
Also report the raw internal length of the independent 1.0 m reference and the resulting scale factor.
Live performance
| Window config | Inference p50/p95 | Alignment p50/p95 | Update interval p50/p95 | Peak memory | Avg/peak power | Max temp |
|---|---|---|---|---|---|---|
| Locked configuration |
Sustained run
| Duration | Accepted windows | Rejected windows | Final points | Peak memory | Max p95 update interval | Max temp | Pass |
|---|---|---|---|---|---|---|---|
| 15 min |
Required figures
All report evidence must be generated by this project. External demonstrations in Final Demo Target and Reference Systems are presentation references only and must not be used as result figures.
| Figure | Required view or comparison | Required annotations and evidence |
|---|---|---|
| Processed RGB, relative depth, and confidence | The same pixels from one accepted window, shown side by side | Window ID, frame ID, processed resolution, confidence threshold, and shared color scale where applicable |
| Single-window cloud with predicted cameras | One oblique view that makes room structure and all window cameras visible | Window ID, retained-point count, internal normalization, and coordinate axes |
| Two-window overlap before and after Sim(3) | Matched viewpoints with identical crop, camera, point size, and colors | Source/target IDs, shared frames, correspondences, inlier ratio, residual, and estimated scale |
| Final bounded colored cloud and trajectory | One overview plus one close structural view of the doorway, desk, and wall corner | Accepted/rejected windows, final point count, bound events, and relative or metric units |
| Independent scale reference | Source photograph and reconstructed endpoints | Physical 1.0 m length, selected endpoints, raw internal length, and resulting scale factor |
| Doorway width and height | Source photograph and matched reconstruction view | Reference and reconstructed endpoints, dimensions, absolute error, and percentage error |
| Registration outcomes | Accepted and rejected counts grouped by reason | All attempted windows remain in the denominator |
| Window-stage latency | Distribution or timeline for capture, inference, alignment, mapping, and display | p50, p95, units, run duration, and locked window configuration |
| Fifteen-minute telemetry | Synchronized time series for memory, point count, power, temperature, and accepted-update interval | Map-bound events, rejected windows, gaps, throttling, and failure markers |
Use project-generated screenshots or plots at readable resolution. Preserve the original artifact or log behind every figure, and record the run ID in the caption so the evidence can be reproduced.
Reporting conventions
- State whether the primary or fallback window was used
- Report raw relative and metric-scaled results separately
- Use p50 and p95 for latency and update interval
- Preserve rejected attempts in all denominators
- Label manually selected endpoints
- Report all map consolidation and point-cap events
- Distinguish preview frame rate from reconstruction update rate
Interpretation
Address these questions directly:
- Did DA3 Small fit the target Jetson stack?
- Was geometry coherent inside individual windows?
- Were overlapping windows joined reliably?
- What metric error remained after one independent scale alignment?
- Was the chunked live update rate convincing and stable?
Do not equate low overlap residual with ground-truth pose accuracy. Do not describe the scaled map as survey-grade. State the restricted scene and motion conditions.