Computer based object detection within a video or image
Described herein are software and systems for analyzing videos and/or images. Software and systems described herein are configured in different embodiments to carry out different types of analyses. For example, in some embodiments, software and systems described herein are configured to locate an object of interest within a video and/or image.
1. A computer-implemented method for evaluating a joint of an individual, the method comprising:
(a) inputting a video of the individual comprising a plurality of frames into a software module, wherein the software module comprises Monte Carlo sampling;
(b) identifying the joint of the individual within the video with the software module;
(c) locating the joint at one or more times in the video with the software module, wherein locating the joint comprises using the Monte Carlo sampling to generate sample locations of the joint, wherein the Monte Carlo sampling comprises
(i) generating draws from a probability distribution characterizing the joint, wherein the probability distribution is parameterized by at least a frame of the plurality of frames; and
(ii) calculating one or more expectation values characterizing the joint from the draws from the probability distribution;
(d) measuring an angle of the joint at one or more times in the video with the software module; and
(e) assessing the joint based on the measurement of the previous step.
2. The method of claim 1 , wherein the joint comprises a shoulder, elbow, hip, knee, or ankle.
3. The method of claim 1 , wherein the probability distribution comprises a categorical distribution.
4. The method of claim 1 , wherein the one or more expectation values comprises a mean angle of the joint.
5. The method of claim 1 , wherein the one or more expectation values comprises a mean angle of the joint and a variance of the joint.
6. The method of claim 1 , further comprising generating a function describing the angle of the joint over time with the software module.
7. The method of claim 6 , wherein the software module comprises Bayesian analysis, and wherein the Bayesian analysis generates the function describing the angle of the joint over time.
8. The method of claim 7 , wherein the Bayesian analysis comprises Gaussian process regression.
9. The method of claim 8 , wherein inputs to the Gaussian process regression comprise the angle of the joint.
10. The method of claim 1 , wherein the software module comprises a deep neural network.
11. The method of claim 10 , wherein the deep neural network comprises at least one of VGG-19, ResNet, Inception, and MobileNet.
12. A computer-based system for evaluating a joint of an individual, the system comprising:
(a) a processor;
(b) a non-transitory computer-readable medium comprising a computer program configured to direct the processor to:
(i) input a video of the individual comprising a plurality of frames into a software module, wherein the software module comprises Monte Carlo sampling;
(ii) identify the joint of the individual within the video with the software module;
(iii) locate the joint at one or more times in the video with the software module, wherein the Monte Carlo sampling is used to locate the joint by generating sample locations of the joint, wherein the Monte Carlo sampling comprises
(1) generating draws from a probability distribution characterizing the joint, wherein the probability distribution is parameterized by at least a frame of the plurality of frames; and
(2) calculating one or more expectation values characterizing the joint from the draws from the probability distribution; and
(iv) measure an angle of the joint at one or more times in the video with the software module.
13. The system of claim 12 , wherein the measurement of the angle of the joint is assessed by a healthcare provider.
14. The system of claim 12 , wherein the joint comprises a shoulder, elbow, hip, knee, or ankle.
15. The system of claim 12 , wherein the probability distribution comprises a categorical distribution.
16. The system of claim 12 , wherein the one or more expectation values comprises a mean angle of the joint.
17. The system of claim 12 , wherein the one or more expectation values comprises a mean angle of the joint and a variance of the joint.
18. The system of claim 12 , wherein the computer program further directs the processor to generate a function describing the angle of the joint over time.
19. The system of claim 18 , wherein the function is generated using Bayesian analysis.
20. The system of claim 19 , wherein the Bayesian analysis comprises Gaussian process regression.
21. The system of claim 20 , wherein inputs to the Gaussian process regression comprise the angle of the joint.
22. The system of claim 12 , wherein the software module comprises a deep neural network.
23. The system of claim 22 , wherein the deep neural network comprises at least one of VGG-19, ResNet, Inception, and MobileNet.