A Deep-Learning Approach for Marker-less Stride Parameters Analysis with Two Cameras
Bibliographic record
Abstract
Human gait analysis is an essential indicator for physical and neuroglial health of an individual. Recent developments in deep-learning approaches to computer vision make possible new techniques for body segment and joint detection from photos and video frames. In this thesis, we propose a deep learning approach for non-invasive video-based gait analysis using two RGB cameras that would be suitable for routine gait monitoring in senior care and rehabilitation centers. Due to modularity and the low cost of implementation, it is considered an affordable solution for such centers. Furthermore, since the solution does not require any markers or sensors to be worn, it is a pervasive and easy method for daily usage. Our proposed deep-learning approach starts by calibrating both the intrinsic and extrinsic parameters of the cameras. Next, video streams captured from two RGB cameras are used as input, and OpenPose and HyperPose deep-learning frameworks are used to localize the main body key points, including the joints and skeleton based on Body 25 and COCO models, respectively. The 2D parameter outputs from the frameworks are triangulated into 3D vector spaces for further analysis. In order to reduce the noises in our data, we applied median and dual pass butter worth filters to the data. Finally gait parameters has been extracted measured and compared to the manually evaluated ground truth data which has been capture via manual measurement of a domain expert. The approach was evaluated in a laboratory setting similar to an institutional hallway in five types of trials: walking back and forth in a straight line while turning out of frame, walking back and forth in a straight line while turning in frame, circular walking, walking with a cane and a walker. The method brings promising results compared to more expensive and restrictive approaches that use up to 16 cameras and require markers or sensors.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".