Computer vision system for ambient long-term gait assessment

1. A computer-implemented method for gait analysis of a subject comprising:

obtaining visual data from a single frontal view image capture device, the image capture device being positioned in front of at least one walking subject, the image capture device including a camera mounted in an elongate hallway in which the subject can walk towards the camera, said visual data including a frontal view of the subject approaching the camera, the visual data comprising at least two image frames of the at least one walking subject over a period of time walking toward the image capture device, the at least two image frames capturing at least a portion of the gait of the at least one walking subject;

analyzing the visual data to determine the identity of the at least one walking subject;

for each identified walking subject:

segmenting one or more gait cycles of the identified walking subject, the segmenting including detecting within the at least two images body parts as two-dimensional landmarks using a pose estimation algorithm on each of the at least two frames, generating a joint model depicting the location of the at least one joint in each of the at least two frames, and using the joint model to segment the one or more gait cycles;

detecting at least one gait feature in the one or more gait cycles;

comparing the at least one gait feature to at least one of the following to detect an abnormal gait: a threshold value or a history of gait cycles of the identified walking subject;

repeating the foregoing steps over a period of time to generate a gait profile for each identified walking subject; and

analyzing the gait profile to detect at least one trend in at least one gait cycle over time.

2. The computer-implemented method for gait analysis of a subject as set forth in claim 1 , wherein the at least one trend includes a change in stride length, duration or cadence.

3. The computer-implemented method for gait analysis of a subject as set forth in claim 1 , further comprising, prior to generating the joint model, estimating a three-dimensional shape of the subject using the two-dimensional landmarks, and estimating the at least one joint location based on the three-dimensional shape.

4. The computer-implemented method for gait analysis of a subject as set forth in claim 1 , wherein the joint model includes a deformable parts model.

5. The computer-implemented method for gait analysis of a subject as set forth in claim 1 , wherein the at least one joint includes an ankle, a knee, a hip, or other joint.

6. The computer-implemented method for gait analysis of a subject as set forth in claim 5 , wherein the gait cycle includes a distance between two consecutive peaks in a trajectory of a joint.

7. The computer-implemented method for gait analysis of a subject as set forth in claim 5 , wherein the gait cycle includes a distance between consecutive peaks in an angle of a joint or body part.

8. A system for gait analysis of a subject comprising:

an image capture device operatively coupled to a data processing device and positioned in front of or behind a region interest wherein a plurality of subjects periodically walk, the image capture device including a camera mounted in an elongate hallway in which the plurality of subjects can walk towards the camera, said visual data including a frontal view of the subject approaching the camera;

a processor-usable medium embodying computer code, said processor-usable medium being coupled to said data processing device, said computer code comprising instructions executable by said data processing device and configured for:

monitoring a region of interest with the image capture device wherein a plurality of subjects periodically walk;

detecting activity within the region of interest corresponding to at least one walking subject;

obtaining visual data from the image capture device, the image capture device being positioned in front of at least one walking subject, the visual data comprising at least two image frames of the at least one walking subject over a period of time walking toward the image capture device, the at least two image frames capturing at least a portion of the gait of the at least one walking subject;

analyzing the visual data to determine the identity of the at least one walking subject;

for each identified walking subject:

segmenting one or more gait cycles of the identified walking subject, the segmenting including detecting within the at least two images body parts as two-dimensional landmarks using a pose estimation algorithm on each of the at least two frames, generating a joint model depicting the location of the at least one joint in each of the at least two frames, and using the joint model to segment the one or more gait cycles;

detecting at least one gait feature in the one or more gait cycles;

comparing the at least one gait feature to at least one of the following to detect an abnormal gait: a threshold value or a history of gait cycles of the identified walking subject;

repeating the foregoing steps over a period of time to generate a gait profile for each identified walking subject.

9. The system for gait analysis as set forth in claim 8 , further comprising analyzing the gait profile to detect at least one trend in at least one gait cycle.

10. The system for gait analysis as set forth in claim 9 , wherein the at least one trend includes a change in gait length, duration or angle.

11. The system for gait analysis as set forth in claim 8 , further comprising, prior to generating the joint model, estimating a three-dimensional shape of the subject using the two-dimensional landmarks, and estimating the at least one joint location based on the three-dimensional shape.

12. The system for gait analysis as set forth in claim 8 , wherein the joint model includes a deformable parts model.

13. The system for gait analysis as set forth in claim 8 , wherein the at least one joint includes an ankle, a knee, a hip, or other joint.

14. The system for gait analysis as set forth in claim 13 , wherein the gait cycle includes a distance between two consecutive peaks in a trajectory of a joint.

15. The system for gait analysis as set forth in claim 13 , wherein the gait cycle includes a distance between consecutive peaks in an angle of a joint or body part.

16. A non-transitory computer-usable medium for gait analysis of a subject, said computer-usable medium embodying a computer program code, said computer program code comprising computer executable instructions configured for:

obtaining visual data from the image capture device, the image capture device being positioned in front of at least one walking subject, the image capture device including a camera mounted in an elongate hallway in which the plurality of subjects can walk towards the camera, said visual data including a frontal view of the subject approaching the camera, the visual data comprising at least two image frames of the at least one walking subject over a period of time walking toward the image capture device, the at least two image frames capturing at least a portion of the gait of the at least one walking subject;

analyzing the visual data to determine the identity of the at least one walking subject;

for each identified walking subject:

segmenting one or more gait cycles of the identified walking subject, the segmenting including detecting within the at least two images body parts as two-dimensional landmarks using a pose estimation algorithm on each of the at least two frames, generating a joint model depicting the location of the at least one joint in each of the at least two frames, and using the joint model to segment the one or more gait cycles;

detecting at least one gait feature in the one or more gait cycles;

comparing the at least one gait feature to at least one of the following to detect an abnormal gait: a threshold value or a history of gait cycles of the identified walking subject;

repeating the foregoing steps over a period of time to generate a gait profile for each identified walking subject.