Motion Coach: User Feedback

JavaScript isn't enabled in your browser, so this file can't be opened. Enable and reload.

Feedback regarding the core technology stack (MediaPipe + Gemini 3), the "Hybrid Neuro-Symbolic" architecture, and the overall user experience.

1. Rep Counting Accuracy *

Purpose: To validate the sensitivity thresholds in the RepetitionCounter.

Perfect – It caught every rep exactly.

Missed Reps – It frequently failed to count valid reps (Undercounting).

Ghost Reps – It counted reps when I wasn't moving (Overcounting).

Drifting – It stopped counting halfway through the set.

2. Live Form Feedback (The "Lanista") *

Purpose: To test the analyzeForm heuristics and the audio cue system.

Yes, the cues were timely and fixed my mistake immediately.

The cues were correct, but slightly delayed.

No, the cues were annoying or false (e.g., told me to "Straighten Back" when it was already straight).

I didn't hear/receive any audio cues.

3. Exercise Recognition (Auto-Detect) *

Purpose: To validate the Gemini 3 Flash calibration phase and locking logic.

Instantly – Locked on within the first 1-2 reps.

Slow – Took 5+ seconds or multiple reps to lock on.

Incorrect – It identified the wrong exercise (e.g., confused Squat for Deadlift).

Failed – I had to manually select the exercise from the list.

4. Post-Workout Analysis Depth *

Purpose: To measure the value of the Gemini analyzeExerciseFrame API call.

Game Changer – The biomechanical insights were specific and new to me.

Good – Useful summary, but I already knew most of it.

Generic – The feedback felt like a standard textbook definition, not specific to my video.

Inaccurate – The analysis described mistakes I definitely did not make.

5. Privacy Perception *

Purpose: To validate the WebGL background stripping and Face Masking.

Completely Secure – I felt comfortable using this in my home/gym.

Neutral – I didn't really notice or care.

Distracting – The masks made it hard to see my own form on screen.

Unsafe – I still don't trust the camera being on.

6. System Performance & Latency *

Purpose: To check the weight of the MediaPipe + React render loop.

Smooth – 60fps, no lag.

Playable – Minor stuttering, but usable.

Hot/Heavy – My phone got very hot or battery drained rapidly.

Unusable – The camera feed was extremely laggy or froze.

7. Retention & Utility *

Purpose: To determine product-market fit.

Yes, for every workout (My Primary Tracker).

Occasionally, just for technique audits (Weekly/Monthly).

Maybe, but only if it supported more exercises/gym equipment.

No, setting up the camera/phone is too much friction for regular use.

Are there additional exercises you would like to see in the standard library for rep counting?

Any additional feedback you would like to share?

Submit

Clear form

Never submit passwords through Google Forms.

This content is neither created nor endorsed by Google. - Terms of Service - Privacy Policy

Does this form look suspicious? Report

Forms