This disclosure describes systems, methods, and devices related to automatic annotation. A device may capture data associated with an image comprising an object. The device may acquire input data associated with the object. The device may estimate a plurality of points within a frame of the image, wherein the plurality of point constitute a 3D bounding to around the object. The device may transform the plurality of points to two or more 2D points. The device may construct a bounding box that encapsulates the object using the two or more 2D points. The device may create a segmentation mask of the object using morphological techniques. The device may perform annotation based on the segmentation mask.

The present invention provides a professional dance evaluation method for implementing Human Pose Estimation based on Deep Transfer Learning. First of all, the Transfer Learning principle of deep learning is combined with the pose features of professional dance training to build a Human Pose Estimation model. Afterwards, the video of demonstration dancing actions is collected and imported into the Human Pose Estimation model to obtain the time-dependent body keypoint data as the reference standard for evaluation. Finally, the video of the examinee's dancing actions is collected and imported into the Human Pose Estimation model to obtain the body keypoint data of the examinee's dancing actions, the similarity between it and the reference standard for evaluation is used for evaluating the standard level of dancing pose.

A method of executing a gesture command includes identifying a hand centroid of a hand. The method also includes identifying a first finger tip of a first finger on the hand. The method further includes identifying a thumb tip of a thumb on the hand. Moreover, the method includes determining a surface normal relationship between the hand centroid, the first finger tip, and the thumb tip.

This invention discloses a practical method for landslide detection in large space, which comprises the following steps: image synthesis, ice and snow detection, removal of non-potential landslide area, detection of potential landslide area, feature calculation, landslide detection model construction and precision validation; this invention avoids radiometric correction and outlier by detecting landslide from synthetic image. That guarantees practical applicability of the proposal. Firstly, detecting potential landslides can avoid the imbalanced sample distribution issue between background objects and landslides when training the landslide detection model. The landslide is further detected by building a random forest model based on the spectral features and textural features of potential landslide pixels in different neighboring time domains. It fully considers the changes of objects in different time domains, and lays a foundation for efficient landslide extraction. This model is relatively reliable and practical for automatically detecting landslide from large-scale images.

This invention discloses a practical method for landslide detection in large space, which comprises the following steps: image synthesis, ice and snow detection, removal of non-potential landslide area, detection of potential landslide area, feature calculation, landslide detection model construction and precision validation; this invention avoids radiometric correction and outlier by detecting landslide from synthetic image. That guarantees practical applicability of the proposal. Firstly, detecting potential landslides can avoid the imbalanced sample distribution issue between background objects and landslides when training the landslide detection model. The landslide is further detected by building a random forest model based on the spectral features and textural features of potential landslide pixels in different neighboring time domains. It fully considers the changes of objects in different time domains, and lays a foundation for efficient landslide extraction. This model is relatively reliable and practical for automatically detecting landslide from large-scale images.

A collation device includes a processor configured to, by executing a program: (a) acquire a photographed image including a collation area provided on a printing substrate having unevenness, (b) simultaneously execute smoothing processing and shading difference enhancement processing on the photographed image, and (c) detect the collation area based on an image obtained by simultaneously executing the smoothing processing and the shading difference enhancement processing.

An information processing device obtains each piece of image data captured within a period of time from entering until exiting of a person at a store. The information processing device identifies joint positions of a skeleton related to the person by analyzing each piece of the image data. The information processing device identifies, as an action which indicates a degree of interest of the person in the product, an action performed by the person to a product in the store from the entering until the exiting, on a basis of the joint positions of the skeleton. The information processing device generates a detection rule that correlates the identified action and the product with each other.

An information processing apparatus detects a person and a commodity product from image data. The information processing apparatus acquires, from the image data, a position of a skeleton of the person included in skeleton information on the detected person. The information processing apparatus specifies, based on the position of the skeleton of the person, a behavior of the person exhibiting with respect to the commodity product. The information processing apparatus specifies, based on the specified behavior of the person exhibiting with respect to the commodity product, a combination of an attribute of the commodity product and a degree of interest in the commodity product.

Systems and methods are disclosed for generating and providing guided practices and coaching feedback (e.g., illustrated in 3D images, video, or audio) that allow golfers to follow along to improve their physical swing in a digital environment. The feedback may be provided with a 3D avatar using a biomechanical analysis of observed actions with a focus on representing actions through computer-generated 3D avatars. Physical quantities of biomechanical actions can be measured from the observations, and the system can analyze these values, compare them to target or optimal values, and use the observations and known biomechanical capabilities to generate the guided practices and coaching feedback.

Systems and methods are disclosed for generating and providing guided practices and coaching feedback (e.g., illustrated in 3D images, video, or audio) that allow golfers to follow along to improve their physical swing in a digital environment. The feedback may be provided with a 3D avatar using a biomechanical analysis of observed actions with a focus on representing actions through computer-generated 3D avatars. Physical quantities of biomechanical actions can be measured from the observations, and the system can analyze these values, compare them to target or optimal values, and use the observations and known biomechanical capabilities to generate the guided practices and coaching feedback.