Generating orthotic product recommendations
A method to generate orthotic product recommendations for users includes receiving interactions from a user including identification of one or more extremity areas of the user's body, the one or more extremity areas relevant to selection of one or more orthotic products. On-screen cues and interactions are provided to assist in positioning of the one or more extremity areas relative to one or more imaging sensors. Scan data is received from the one or more imaging sensors. A model of the one or more extremity areas is generated based on the scan data including estimating one or more complete circumferences of the one or more extremity areas. One or more orthotic products are identified based on a comparison of the model of the one or more extremity areas and one or more factors associated with the user and a plurality of orthotic products. A recommendation is provided that includes the one or more orthotic products.
1. A computer-implemented method comprising:
receiving interactions from a user comprising identification of one or more extremity areas of the user's body, the one or more extremity areas relevant to selection of one or more orthotic products;
providing on-screen cues and interactions to assist in positioning of the one or more extremity areas relative to one or more imaging sensors;
receiving scan data from the one or more imaging sensors, the scan data comprising images of a front of the user;
generating a model of the one or more extremity areas based on the scan data comprising estimating one or more complete circumferences of the one or more extremity areas;
identifying the one or more orthotic products based on a comparison of the model of the one or more extremity areas and one or more factors associated with the user and a plurality of orthotic products; and
providing a recommendation that includes the one or more orthotic products.
2. The method of claim 1 , wherein estimating the one or more complete circumferences of the one or more extremity areas comprises determining a shape of the one or more extremity areas that is not visible from the images.
3. The method of claim 1 , wherein the one or more factors comprises a body mass index of the user.
4. The method of claim 1 , wherein providing the on-screen cues and interactions comprises:
receiving the scan data from the one or more imaging sensors, wherein the scan data is of a leg that includes a knee;
determining whether the leg is correct;
responsive to determining that the leg is correct, determining whether the knee is visible; and
responsive to the knee being visible, determining whether the leg is in a first position.
5. The method of claim 4 , further comprising responsive to determining that the leg is incorrect, the knee is not visible, or the leg is in a second position, instructing the user to make a modification.
6. The method of claim 4 , wherein the knee is not visible because a garment obstructs part of the knee.
7. The method of claim 1 , wherein generating the model of the one or more extremity areas based upon the scan data further comprises:
performing point-cloud segmentation to divide the scan data into point clusters;
identifying a point-cluster centroid for each of the point clusters;
identifying locations of joints in a leg of the user based on the point-cloud centroids;
generating a two-dimensional depth map of the leg based on the locations of the joints;
generating a point cloud of the leg based on the two-dimensional depth map;
determining one or more partial circumferences that correspond to one or more points of the leg from the point cloud; and
determining the one or more complete circumferences based on the one or more partial circumferences.
8. The method of claim 1 , wherein the one or more complete circumferences have an ovoid shape.
9. The method of claim 1 , wherein the recommendation is provided to an email address of the user.