Sap Se

Raghavendra D

Sivakumar N

Vijay G

Praveenkumar A K

Pratik Shenoy

Kishan Kumar Kedia

Mithilesh Kumar Singh

Satyadeep Dey

Philippe Long

Anubhav Sadana

Deepak Pandian

Disclosed herein are system, method, and computer program product embodiments for surface automation in black box environments. An embodiment operates by determining scenarios of an application for automation; detecting the scenario during an execution of an application; capturing and storing one or more user interface screenshots of the scenario; identifying and storing user interface information from the user interface screenshot; implementing a sequential set of instructions comprising at least one textual element detection technique and at least one non-textual element detection technique; and executing the sequential set of instructions.

Surface automation in black box environments

In an example, a computerized neural fabric is created by representing each pattern of learned weights of one or more machine learning algorithm-trained models specifying a specific set of products as a column in the computerized neural fabric, each pattern comprising one or more clusters representing combinations of convolutional filters, each cluster learning low level features and sending output via a vertical flow up the corresponding column to a final classification within the corresponding pattern. One or more potential lateral flows between patterns in the computerized neural fabrics is dynamically determined based on resemblance of a new product in a candidate image to the specific sets of products in each of the patterns and identifying possible mutations of the patterns based on the resemblance. Then, one of the one or more potential lateral flows is selected as a new model.

Dynamic self-learning system

In an example, a first machine learning algorithm is used to train a smart contour model to identify contours of product shapes in input images and to identify backgrounds in the input images. A second machine learning algorithm is used to train a plurality of shape-specific classification models to output identifications of products in input images. A candidate image of one or more products is obtained. The candidate image is passed to the smart contour model, obtaining output of one or more tags identifying product contours in the candidate image. The candidate image and the one or more tags are passed to an ultra-large scale multi-hierarchy classification system to identify one or more classification models for one or more individual product shapes in the candidate image. The one or more classification models are used to distinguish between one or more products and one or more unknown products in the image.

Computer vision architecture with machine learned image recognition models

Growing community of inventors

Bengaluru, India

Raghavendra D