Cross-platform client
Flutter handles identity, upload, live job state, reports, and deletion across mobile and web.
Built for accountable production
Turn a phone video into a traceable tennis report.
AceSense analyses uploaded tennis footage for shot events, court-relative patterns, pose-derived observations, heatmaps, and frame-linked review on iOS, Android, and web.
Repository-backed product facts · reviewed 26 August 2026
Architecture and value
Use the original video as evidence, preserve confidence and unavailable markers, and choose one coaching priority rather than treating a score as medical or officiating truth.
Owner-scoped control plane
Firebase owns owner-scoped upload authorization, queue state, billing, dispatch, completion, and deletion.
Failure-aware GPU pipeline
Disposable GPU workers run a staged vision pipeline and publish immutable generation-scoped artifacts.
Compare operating models
Expand the matrix. Choose by boundary and workflow, not slogans.
AceSense ↔ SwingVision · current platform and plan terms must be verified before purchase
| This system | Alternative · SwingVision | |
|---|---|---|
| Primary job | Post-session shot, movement, pose, and frame-linked review | Tennis video analysis with prominent live and Apple-device workflows |
| Deployment boundary | iOS, Android, and web client with asynchronous GPU inference | Verify current supported devices and processing model |
| Evidence model | Confidence-aware events linked back to source frames | Match statistics and video review; verify export methodology |
| Operator control | Owner-scoped jobs, immutable artifacts, explicit deletion | Consumer application workflow and account controls |
| Integration surface | Mobile/web app, REST v1, and MCP | Mobile app and current vendor integrations |
Technical invariants
Owner isolation
Every job, upload, result, and deletion is owner-scoped.
Every job, upload, result, and deletion is owner-scoped.
Atomic publication
A canonical result pointer changes only after the complete generation is durable.
A canonical result pointer changes only after the complete generation is durable.
Measurement honesty
Unknown or unavailable evidence remains unknown; the product does not invent contact, 3D, or line calls.
Unknown or unavailable evidence remains unknown; the product does not invent contact, 3D, or line calls.
Bounded video access
Short-lived capabilities, deletion paths, and explicit EU-to-US inference transfer boundaries constrain video handling.
Short-lived capabilities, deletion paths, and explicit EU-to-US inference transfer boundaries constrain video handling.
Transparent access
Free
€0 / month
Three analyses per month with the complete report surface.
Pro
€19 / month
Higher analysis limits for an individual player.
Team
€49 / month
Pro features for up to ten players and coach workflows.
Current status. The analysis pipeline and public API are production surfaces; Stripe remains in test mode and ordinary monocular video is not an official line-calling or medical measurement system.
Frequently asked questions
What is the safest way to evaluate it?
Start with the architecture, reproduce the documented workflow, and verify the boundary against your own threat model.
Does the comparison include live third-party pricing?
No. Mutable vendor pricing is deliberately excluded. Verify current vendor terms before a purchase decision.
Where are limitations documented?
The status note, architecture page, and engineering articles state the current boundary and failure behaviour.