FIRST WAVE

The best partof your storyisn't behind you.

At 20. At 45. At 65. There are still first times: a first wave, a first marathon, a first yoga class, a better version of yourself.

7+Engine profiles
33Key points analysed
4INPI patent filings
SCROLL
Why we exist

Performance fades.
Progress doesn't.

BiokinetIQ was created for those who believe their story is still being written. Not to become the best. But to become better. For athletes, for coaches, and for anyone who refuses to become a spectator of their own life.

A journey that respects time, the body, experience and transmission.

More than yesterday.
Less than tomorrow.
BIOKINETIQ · LIFELONG PROGRESSION
Every movement tells a story

Fear.
Commitment.
Joy.

Every sport carries the same passage: trying, learning, building confidence, transmitting, beginning again. BiokinetIQ keeps that story readable session after session.

The knowledge of coaches

01

Coaches are memory keepers

A look, a correction, the right words at the right moment can change a trajectory. That knowledge should not disappear after the session.

02

Studio preserves transmission

Feedback, corrections, video blocks, analysis and progression history become an active memory of coaching.

03

The coach remains a transmitter

BiokinetIQ does not turn coaches into content creators. It helps their knowledge travel, adapt and stay alive.

The student's journey

01

First attempt

The beginning is rarely perfect. It matters because it shows where the story starts.

02

Learning and confidence

Session after session, the body understands. The movement becomes clearer, steadier, more personal.

03

Lifelong evolution

Mastery is not an ending. It is a passage toward the next version of yourself.

More than yesterday · Less than tomorrow
One thread across every sport

Not performance.
Progress.

A wave, a posture, a repetition, a swing, a route, a run: every discipline changes the movement, but not the human need to move forward.

MaxWaxINTEGRATION

Surf & Surfskate

Reading the fear of the take-off, the commitment in the turn and the joy of a wave better understood.

AVAILABLE

Yoga

Turning patience into signal: holding, breathing, adjusting, returning to a body that learns.

AVAILABLE

Fitness

Not just sweating. Understanding what improves, what fatigues, and what deserves protection.

ROADMAP

Golf

Seeing a movement built slowly: balance, rotation, rhythm, intention.

ROADMAP

Climbing

Understanding the relationship between fear, decision, balance and commitment in movement.

ROADMAP

Running

Following what changes when the body learns to last: cadence, foot strike, fatigue and confidence.

INTEGRATION

Surfskate

Repeating the feeling out of the water: compression, rotation, rebound, confidence.

SELECT A SPORT

The four product layers

Android.
Studio. API.
ML engine.

Students practice on mobile. Coaches build and annotate in Studio. Partners integrate the API. The ML engine generates usable session debriefs.

SESSION.LIVE
STUDENTS: 12ACTIVE
COACH
81
GROUP AVG
pts
6
CORRECTIONS
students
4
EXCELLENCE
students
COACH.STUDIOACTIVE
A01
94
A02
78
A03
88
A04
62
A05
91
A06
85
A07
71
A08
97
A09
83
A10
76
A11
55
A12
90
ASYMMETRY.HEATMAP
LOWHIGH ASYMM.
Student app
Onboarding, sessions, camera practice, feedback, progression and opportunistic video analysis.
Coach Studio
Video upload, blocks, annotations, biomechanical focus, publishing and share links.
Developer API
Analysis jobs, engine profiles, biomechanical JSON and webhooks for third-party apps.
ML coaching engine
Post-session debriefs from scores, dimensions, strong blocks and improvement axes.
BiokinetIQ engine

Biomechanical analysis
serving the decision

The engine turns video into blocks, signatures, joint loads, scores and confidence levels to help the preparation layer decide what should happen next.

VIDEO
Phone · Camera · Tablet
SKELETON
33 key points · real time
SEGMENTS
Movements · Reps · Phases
SIGNATURE
Blocks · Loads · Intentions
MATCHING
Profile · Goal · Constraints
92
ACTION
Recommend · Adapt · Replace
CORE ENGINE

Mobile and server analysis

Mobile real-time analysis supports practice. Deeper analysis runs through the API and video pipeline when the context requires it.

Android + API
product surfaces
TRACKING

Every measure keeps its confidence level

The engine accounts for signal quality, landmarks and video conditions instead of presenting the score as an absolute truth.

Signal
signal quality
PREPARATION

The session adapts to today's profile

Goal, level, pain, fatigue, pedagogical preferences and history guide the recommendation and the active timeline.

Adaptatif
adaptive decision
AI COACHING

A readable post-session debrief

The coaching model turns scores, strong blocks, improvement axes and perceived effort into useful student-facing feedback.

FR/EN
FR/EN feedback
Intellectual property

Four patent filings
with INPI

The filings cover four complementary layers: making measurement more robust, preserving tracking continuity, interpreting movement through motor profiles and building compatibility between teaching and learner profiles.

INPI FILING
BKIQ-01

Robust biomechanical measurement

Adaptive filtering of biomechanical features based on detection confidence and an expert statistical reference model.

ADAPTIVE FILTERING · EXPERT MODEL
T₁p=0.97T₂p=0.82T₃p=0.95T₄p=0.61P(x_t | z_{1:t}) ∝ P(z_t | x_t) · ∫ P(x_t | x_{t-1}) P(x_{t-1} | z_{1:t-1}) dx_{t-1}
Positions, angles and velocities associated with a confidence index
Regularisation of unreliable components toward an expert reference
More interpretable comparison under degraded video conditions
INPI FILING
CVTE

Identity and observation continuity

Maintaining a probabilistic belief state through visibility loss, identity conflict and non-stationary scenes.

DUAL ACCUMULATION · HYSTERESIS
motionpose_okconf>0.8vis<0.3t>T_maxvis>0.6match>θfailIDLEDETECTINITTRACKOCCLUDEDRE-IDP(id|obs) = Σ P(id|s) · P(s|obs)
Identity substitution conditioned by score and history
Eleven tracking states, from LOCKED to RELEASED
Possible control of a motorised stabilisation device
INPI FILING
PM-A

Adaptive motor profile

A motor profile governs segmentation, states, body-segment weights, features and comparison rules.

STATES · WEIGHTS · SIGNATURE
P = (E, W, F_prio, T_rules, C_rules)TRUNK0.92PELVIS0.86LEGS0.74ARMS0.48
The same signal can produce different readings depending on the profile
Body segments weighted by biomechanical relevance
Temporal kinematic signature governed by the profile
INPI FILING
PCS

Teaching-learner compatibility

Building homologous spaces to compare coaching content and learner profiles across biomechanical and pedagogical dimensions.

CONTENT · PROFILE · COMPATIBILITY
SESSION SIGNATURELEARNER PROFILEBPBPPCS = w_b · sim(B) + w_p · sim(P)
Digital session representation from video and audio
Learner profile combining abilities, limitations and preferences
Compatibility score computed in shared spaces
P(id_t | z_{1:t}) ∝ ∫ P(z_t | s_t) · P(s_t | s_{t-1}) ds
θ̂ = argmax_θ Σ log P(obs_i | skeleton(θ))
σ²_joint = f(conf_i, σ²_model, dt)
score = w_T · φ_T(ψ) / Z_sport
The foundations

A technical foundation
for a useful decision

BiokinetIQ combines computer vision, simplified biomechanics, temporal modelling, student feedback and coach data to guide progression.

KINEMATIC SIGNAL · PHASES DETECTED IN REAL TIME
01
EXPERTISE

Reading your body with precision

Extract usable landmarks from mobile, coach or API videos, with an associated quality level.

Pose estimationOptical flowDepth inferenceOcclusion handling
02
EXPERTISE

Understanding what you're really doing

Turn coordinates into angles, symmetry, amplitude, stability, joint loads and movement signatures.

Inverse kinematicsJoint torquesSegment dynamicsMarkerless
03
EXPERTISE

An honest measurement

Each score must remain interpretable: signal quality, profile limits and execution context matter in the decision.

Bayesian filteringParticle filtersUncertainty quantificationCalibration
04
EXPERTISE

A journey that learns

Post-session feedback feeds preferred difficulty, intensity, pedagogy and joint stresses to avoid.

FeedbackAdaptive timelineMatchingAccuracyLearning profile
05
EXPERTISE

Adapted product surfaces

Android, Studio, API and ML engine do not share the same UX, but they share signatures, blocks, scores and profiles.

Model compressionQuantisationPruningMobile deployment
06
EXPERTISE

Understanding the meaning of movement

Not just coordinates — understanding what you're doing, why, and how to improve across supported profiles and levels.

Temporal modellingAction qualityRepresentation learningPhase segmentation

VISION · BIOMECHANICS · STUDENT PROFILE · ADAPTIVE DECISION

For developers

Integrate analysis and sport decisions.

The developer API creates video analysis jobs, returns versioned biomechanical JSON and can notify your app by webhook. The Android mobile engine complements real-time use cases.

Request API access
API
Video analysis jobs
v2
Versioned biomechanical JSON
7+
Specialised engine profiles
HMAC
Analysis completion webhooks
The future

Your best session
may still be ahead.

No matter your age. No matter where you start. Your story isn't over. More than yesterday. Less than tomorrow.

BIOKINETIQ.TECHSTUDIO.BIOKINETIQ.TECHAPP ANDROID
Biokinetiq

© 2025 BiokinetIQ. All rights reserved.

4 PATENT FILINGS WITH INPI
SPORTS PREPARATION · BIOMECHANICS