2-minute overview
Kineologist GAIT · Scientist
Walking biomechanics from motion files to event-locked evidence.
Run the GAIT pipeline once, then inspect IC/TO events, phase-locked comparison, telemetry, reports, plots and technical evidence at the depth the investigation requires.
GAIT workflow
GAIT is easy to run.
Load baseline and subject files, choose pipeline options, run the pipeline, then review the outputs. GAIT keeps the user workflow simple while the backend handles IC/TO event detection, fixed-step crop windows, same-foot baseline logic and walking-specific telemetry.
Load baseline files
Add reference walking files so GAIT can build a comparable walking baseline.
Load subject file
Add the walking file that will be reviewed against the generated reference movement.
Choose options
Select the gait workflow, baseline behaviour, report scope and analysis settings, then run the pipeline.
Detect IC/TO
GAIT detects initial contact and toe-off events to create walking-cycle timing truth.
Create fixed steps
Walking clips are cropped into fixed-step windows using same-foot logic where required.
Pipeline steps
The ten-stage automated GAIT pipeline.
GAIT separates operator work from walking-event evidence generation. The software detects the cycle structure, builds comparable windows, extracts walking telemetry and produces review artifacts.
01
Load baseline files
Add reference walking files so GAIT can build a comparable walking baseline.
02
Load subject file
Add the walking file that will be reviewed against the generated reference movement.
03
Choose options
Select the gait workflow, baseline behaviour, report scope and analysis settings, then run the pipeline.
04
Detect IC/TO
GAIT detects initial contact and toe-off events to create walking-cycle timing truth.
05
Create fixed steps
Walking clips are cropped into fixed-step windows using same-foot logic where required.
06
Build baseline
Reference walking cycles are normalised and averaged into a repeatable comparison authority.
07
Analyse subject
The subject file is processed with the same event and crop logic used by the baseline.
08
Compare evidence
Cycle timing, contact state, cadence, sway, asymmetry and magnitude differences are reviewed.
09
Evaluate rule logic
Walking rules consume event-relative telemetry, apply thresholds and grade review findings.
10
Generate outputs
GAIT writes rule findings, telemetry, reports, plots, thresholds and CSV evidence for review.
Average walking baseline
Create a same-foot walking baseline for repeatable comparison.
GAIT baseline creation is built around walking events rather than pitching phases. Baseline files are event-marked, cropped into fixed-step windows, normalised and averaged so the subject file can be reviewed against a comparable walking reference.
Same-foot anchor
Baseline and subject windows can be anchored around comparable contact events.
Average walking reference
Reference walking files become a repeatable comparison authority.
Fixed-step crop
GAIT uses walking-event windows instead of arbitrary frame ranges.
Comparable outputs
Markers, telemetry and reports stay tied to the same generated baseline logic.
Event review
Initial contact, toe-off and fixed-step windows define the review structure.
The walking cycle is segmented by event markers. IC and TO events create the timing structure for contact state, contact pulse, fixed-step crop and baseline-relative comparison, with detector refinement available after the initial positional/speed pass.
Initial contact
Contact events anchor the start of the walking cycle after positional and speed evidence is reviewed.
Toe-off
Toe-off events structure release and swing timing before post-process refinement.
Contact channels
Contact pulse and contact state remain inspectable as telemetry.
Same-foot windows
Baseline and subject compare from matching step logic.
Phase-locked GAIT comparison
Walking comparison can align around IC/TO events.
GAIT also uses the reusable phase-warp system. Original view preserves native step timing, while Phase-locked view aligns the reference baseline around repeated initial-contact and toe-off anchors so equivalent parts of the walking cycle can be compared.
IC/TO anchors
Initial contact and toe-off events define the walking comparison map.
Original view
Native step timing and asymmetry remain available for review.
Phase-locked view
Equivalent cycle events align for shape, timing and magnitude inspection.
TrackView context
Contact state, contact pulse and telemetry remain tied to backend event truth.
Dual-pass footstep detection
GAIT uses a two-pass precision system for IC/TO events.
The first pass uses local positional foot traces and speed behaviour to find the likely initial-contact and toe-off structure. A second post-process then uses the motion-derived muscle analysis layer to refine those event frames to a higher precision. The result is walking-event timing that is still inspectable as artifacts, but sharpened beyond a single detector signal. The same refined IC/TO event truth also provides the anchors used for GAIT phase-locked comparison.
Pass 1: position and speed
Local foot traces and speed changes establish the first IC/TO event candidates.
Pass 2: muscle proxy refinement
Motion-derived muscle channels refine event timing after the initial detector pass.
Sharper IC/TO timing
Initial contact and toe-off become high-precision review anchors for fixed-step windows.
Artifact-first audit
The refined events still feed telemetry, CSVs, plots and reports as deterministic outputs.
GAIT telemetry
Walking telemetry and cycle evidence tell the movement story.
GAIT creates walking-specific telemetry so analysts can inspect rhythm, cadence, symmetry, contact behaviour, foot clearance, trunk/head sway and upper-body compensation across the walking cycle.
Contact state
Shows when each foot is considered loaded/contacting during the walking cycle.
Contact pulse
Makes discrete IC/TO event behaviour visible as inspectable telemetry.
Foot clearance
Supports review of swing phase and clearance through each step.
Cadence
Summarises rhythm and timing across repeatable walking cycles.
Asymmetry
Highlights left/right and subject-versus-baseline differences.
Trunk and head sway
Shows balance, compensation and upper-body movement behaviour.
Arm cycles
Keeps upper-limb rhythm available during walking review.
IC and cycle timing
Anchors the review to walking events rather than arbitrary frame ranges.
Rule-based analysis
GAIT uses correct step, stride, stance and swing mathematics.
A step is measured from the opposite foot's initial contact to the current foot's initial contact. A stride is measured from one foot's initial contact to its next same-side contact. Stance and swing are calculated inside each valid same-side stride. GAIT also calculates left and right stride variability independently.
True step time
Opposite-foot IC to current-foot IC.
True stride time
Same-foot IC to the next same-foot IC.
Stance and swing
Calculated inside each valid same-side stride.
Per-side variability
Left and right stride sequences are assessed independently.
Walking review
Review walking motion against events, curves and baseline deltas.
The visual review combines 3D motion with event-aware telemetry. A reviewer can inspect how the subject differs from the baseline in timing, magnitude, clearance, sway and rhythm.
Auto baseline generation
Baseline walking files can be event-cropped, normalised and averaged into a walking reference.
Auto phase marker placement
GAIT places walking event markers so review happens around IC/TO and cycle evidence.
Single subject analysis
A walking file can be analysed independently before baseline comparison is introduced.
Subject vs baseline analysis
The subject can be reviewed against the walking baseline for timing, magnitude and rhythm differences.
Coaching layer deltas
Delta outputs help convert baseline comparison into readable review evidence.
Muscle activity visualizer
Motion-derived proxy overlays help inspect walking contribution patterns without claiming EMG.
Motion-derived proxy layer
Cycle-relative muscle proxy sequencing supports walking review.
GAIT uses motion-derived proxy signals to help review major muscle-group sequencing, cycle-relative contribution and baseline differences through the walking cycle. This gives reviewers another way to inspect movement contribution without claiming EMG or direct measured activation.
Cycle-relative proxy
Proxy signals are reviewed around gait cycles rather than arbitrary frames.
Sequencing context
Contribution patterns can be inspected across the walking cycle.
Baseline differences
Proxy contribution can be compared against the selected walking reference.
Clear boundary
This is not EMG or direct measured activation.
Viewport review
Flexible viewport controls keep side-by-side walking inspection readable.
Perspective/front/side views, camera-local pan, orbit/centered modes and compare-offset controls help reviewers inspect walking motion without fighting the camera.
Orbit / centered
Switch camera behaviour to match the review task.
Perspective / front / side
Use standard viewpoints for gait inspection.
Camera-local pan
Move around the walking subject without losing context.
Compare offset
Separate subject and baseline for side-by-side inspection.
Muscle & labelling options
Reviewers can switch muscle overlays and labels while inspecting walking motion beside event and curve evidence.
Measurement quality
Wrap-safe trunk–pelvis movement without false angle jumps.
GAIT uses continuous pelvis-relative thorax orientation rather than unstable raw Euler angles. The quaternion-based method avoids false wrap jumps and summarises trunk–pelvis movement across valid strides. This is a movement-quality proxy, not a balance diagnosis or fall-risk score.
Continuous orientation
Pelvis-relative thorax movement remains stable across angle boundaries.
Stride-aware summary
Movement is summarised across valid walking cycles.
Fewer false jumps
Quaternion continuity avoids raw Euler wrap artifacts.
Clear boundary
The output is a movement-quality descriptor, not a clinical score.
Coach report
Every valid walking take produces a complete Coach profile.
The GAIT Coach report covers capture quality, step symmetry, stride timing, left/right variability, stance and swing, trunk–pelvis movement, baseline comparison, practical review priorities and muscle-proxy highlights. A clean take still produces useful pass-state evidence instead of appearing empty.
GOOD
The available event and movement evidence supports normal interpretation.
LIMITED
The result can be reviewed, but weak or missing evidence should temper conclusions.
INVALID
The take does not support a reliable report and should be repeated or corrected.
Safe near-zero comparison
Subject, baseline and signed physical difference replace misleading extreme percentages.
Muscle report
The GAIT Muscle Report tables cycle-relative proxy contribution.
The Muscle Report supports deeper review of motion-derived proxy sequencing through the walking cycle. It helps reviewers inspect cycle-relative contribution patterns and compare how proxy groups differ from the reference movement. It is not EMG and does not claim direct measured activation.
Cycle-relative proxy
Proxy contribution is reviewed around gait cycles, not arbitrary frames.
Sequencing table
Tables support review of relative order and contribution.
Comparison context
Baseline-relative differences can be inspected across the walking cycle.
Motion-derived
The report does not claim EMG or direct measured activation.
Master report
The GAIT Master Report preserves deeper technical context.
The Master Report is the longer-form review surface for analysts, educators and validation partners who need more context around events, telemetry, baseline comparison, report findings and exported evidence.
Detailed evidence
Keeps more generated context available for review.
Long-form output
Explains the run with deeper language than the Coach Report.
Telemetry context
Contact, clearance, cadence, sway and cycle channels remain inspectable.
Validation support
Designed for facilities and education partners testing the workflow.
Plots
Generated walking plots support technical inspection.
GAIT plot outputs make walking channels available outside the main interface for technical inspection, documentation and review.
Walking channels
Plots expose contact, cadence, sway, clearance and rhythm channels outside the GUI.
IC/TO timing
Events remain visible as timing anchors for technical inspection.
Comparison context
Baseline-relative differences can be reviewed alongside raw curves.
Exportable evidence
Plot outputs support documentation, review packs and validation conversations.
Lexicon Editor
Edit GAIT report language without changing event truth.
Coach and Master wording, evidence templates, review priorities, headings and claim boundaries can be adapted while IC/TO events, gait mathematics, severity and numerical evidence remain stable.
Editable report surfaces
Adapt practical wording, evidence templates and review priorities.
Stable IC/TO truth
Editing language never changes the bilateral gait event stream.
Stable severity
Rules, thresholds and evidence quality remain deterministic.
Five-language workflow
The same semantic report roles remain available across all product languages.
Thresholds
Tunable event and overlay thresholds support expert workflows.
Threshold controls allow reviewers to adapt event markers, overlays and rule sensitivity for a specific movement context, population or facility preference.
Event tuning
Thresholds support review of marker and contact behaviour for a given gait context.
Rule bands
Findings can be tuned for validation work, facility preference or protocol design.
Overlay control
Reviewers can adjust what appears in the timeline and plots.
Deployment ready
The same control pattern can be shaped for custom movement products.
Next step
Review walking data through events, cycles and evidence.
Book a walkthrough of Kineologist|GAIT, test the workflow with walking motion data, or discuss a validation partnership.