PILLAR 01
Gait recognition & biometrics
Published journal research on deep-learning gait recognition with covariates (Springer, Elsevier).
6 papers3 capabilities
Research at GaitAI
Built on apublished research record.
A traceable research foundation spanning gait biometrics, pose-based movement analysis, privacy-aware gait data and edge intelligence.
Human gaitTemporal signalRepresentation
08Surfaced academic papers
01Granted patents
04Research pillars
24Connected product modules
01
The record, drawn
Captured gait, the layered movement engine, and the capabilities the published record backs — with each research area's own records one click away.
02
Four research pillars
PILLAR 02
Pose-based gait analysis
Covariate-invariant gait recognition built on pose features (IET Biometrics).
1 paper2 capabilities
PILLAR 03
Privacy-preserving gait data
Protecting gait datasets inside deep-learning pipelines (IET Biometrics).
1 paper1 capability
PILLAR 04
Edge gait analytics (Patent 402202)
Granted Indian patent: covariate-based gait recognition for edge analytics.
1 granted patent1 capability
03
Evidence map
Which research informs which capability, and which products are built on it.
Select a pillar. The map redraws to that pillar's record, the capabilities it informs and the modules built on them.
Select a research area to trace its evidence path
Gait recognition & biometrics
Published journal research on deep-learning gait recognition with covariates (Springer, Elsevier).
6 research records · 3 capabilities · 15 products
These draw on a broad platform capability this work touches, not on what the record specifically addresses. Listed for traceability — not as evidence for these products.
Research establishes the methodological foundation. Product-specific validation establishes fitness for a particular use.
Explore the full record04
Selected research record
Surfaced academic papers and the granted patent, as published.
- Advancements in Artificial Intelligence for Biometrics: A Deep Dive into Model-based Gait Recognition TechniquesEngineering Applications of Artificial IntelligenceElsevier2024
- Journey into Gait Biometrics: Integrating Deep Learning for Enhanced Pattern RecognitionDigital Signal ProcessingElsevier2024
- Deep learning pipelines for recognition of gait biometrics with covariates: a comprehensive reviewArtificial Intelligence ReviewSpringer2023
- Data Preprocessing and Feature Selection Techniques in Gait Recognition: A Comparative Study of Machine Learning and Deep Learning ApproachesPattern Recognition LettersElsevier2023
- A Covariate-Based Gait Recognition System and Method for Edge Analytics Using Optimized Deep Learning PipelinePatent — Government of IndiaGovernment of India · IP India2022
Showing 5 of 9 records.The full research library →
05
Research journey
A decade of research development, in four steps.
- 2014
Movement research
Academic research begins on human gait analysis for Parkinson's disease and the early prediction of movement-related disorders.
- 2016
Biometrics
The work expands into gait recognition for surveillance and security — gait as a non-contact biometric that works at a distance.
- 2022–24
Published record
8 surfaced academic papers with Springer, Elsevier and Wiley · IET, plus granted Indian patent 402202, informing research into edge gait analytics.
- Today
GaitAI platform
24 product modules across 2 families on one movement-intelligence platform. Product-specific validation is separate from the academic record.
06
How we work
Four commitments that shape how outputs are built.
- 01
Traceable to a signal
Every score is built from named movement features — cadence, symmetry, variability, trajectory — so a reviewer can see what moved the number.
- 02
Peer-reviewed foundations
Recognition, pose and privacy components come from work published with Springer, Elsevier and Wiley · IET; the edge pipeline is covered by a granted patent.
- 03
Framed as decision support
Outputs are assessment, screening and monitoring. No product diagnoses — the design constraint, not a disclaimer added afterwards.
- 04
Stated limitations
Each product's technical view lists what constrains capture quality, what the model does not infer, and where human review is required.
- Privacy by default
- Consent & authority
- Explainable outputs
- No unsupported claims
Research creates the foundation. Validation builds the evidence for use.
GaitAI welcomes research, clinical and technical collaborations that can independently evaluate movement-intelligence systems across mobility, rehabilitation, sports and safety settings.
Collaborations are scoped around a single environment or cohort so findings stay reviewable. Nothing on this page describes a completed study.