Patents by Inventor Eckehard Schmidt
Eckehard Schmidt has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20230196080Abstract: A network output is generated by feeding an input vector to an input layer of a neural network model having a plurality of neurons arranged in a sequence of layers, a plurality of neuron weights, and a plurality of neuron biases. The network output is used to determine an output relevance score. Relevance scores at a last layer of the sequence of layers are generated. Relevance scores are obtained at a first layer of the sequence of layers by reverse propagating the relevance scores generated at the last layer through the sequence of layers other than the last layer using the neuron weights and neuron biases. A feature relevance vector is formed based on the input vector and the relevance scores obtained at the first layer and included in a local explainability dataset, which is then used to generate a local explanation of a prediction of the neural network model.Type: ApplicationFiled: December 17, 2021Publication date: June 22, 2023Applicant: SAP SEInventors: Waqas Ahmad Farooqi, Eckehard Schmidt, Jonas Benedict Grill
-
Publication number: 20230196062Abstract: The layers of a neural network model are traversed in sequence one or more times while generating a plurality of relevance scores each time based on neuron weights and neuron biases of the neuron network model. Each relevance score of the plurality of relevance scores quantifies a relevance of a neuron in a lower layer of the sequence of layers to a higher layer of the sequence of layers. One or more relevance vectors can be populated from the plurality of relevance scores generated at the one or more times. Each of the relevance scores in each relevance vector quantifies a relevance of one of the input features to a task for which the neural network model is trained to perform. An explanation of a behavior of the neural network as a whole is generated based on the one or more relevance vectors.Type: ApplicationFiled: December 17, 2021Publication date: June 22, 2023Applicant: SAP SEInventors: Waqas Ahmad Farooqi, Eckehard Schmidt, Jonas Benedict Grill
-
Patent number: 11416748Abstract: Methods, systems, and computer-readable storage media for providing a binary classifier include receiving a biased dataset, the biased data set including a plurality of records, each record being assigned to a class of a plurality of classes, one class including a majority class, performing data engineering on at least a portion of the biased dataset to provide a revised dataset, providing a trained deep autoencoder (DAE) by training a DAE using only records assigned to the majority class from the revised dataset, the trained DAE including a binary classifier that classifies records into one of the majority class and a minority class, validating the trained DAE using validation data that is based on at least a portion of the biased dataset, and providing the trained DAE for production use within a production system.Type: GrantFiled: December 18, 2019Date of Patent: August 16, 2022Assignee: SAP SEInventors: Ajinkya Patil, Waqas Ahmad Farooqi, Jochim Fibich, Eckehard Schmidt, Michael Jaehnisch
-
Publication number: 20210192354Abstract: Methods, systems, and computer-readable storage media for providing a binary classifier include receiving a biased dataset, the biased data set including a plurality of records, each record being assigned to a class of a plurality of classes, one class including a majority class, performing data engineering on at least a portion of the biased dataset to provide a revised dataset, providing a trained deep autoencoder (DAE) by training a DAE using only records assigned to the majority class from the revised dataset, the trained DAE including a binary classifier that classifies records into one of the majority class and a minority class, validating the trained DAE using validation data that is based on at least a portion of the biased dataset, and providing the trained DAE for production use within a production system.Type: ApplicationFiled: December 18, 2019Publication date: June 24, 2021Inventors: Ajinkya Patil, Waqas Ahmad Farooqi, Jochim Fibich, Eckehard Schmidt, Michael Jaehnisch
-
Patent number: 10289925Abstract: Combined color and depth data for a field of view is received. Thereafter, using at least one bounding polygon algorithm, at least one proposed bounding polygon is defined for the field of view. It can then be determined, using a binary classifier having at least one machine learning model trained using a plurality of images of known objects, whether each proposed bounding polygon encapsulates an object. The image data within each bounding polygon that is determined to encapsulate an object can then be provided to a first object classifier having at least one machine learning model trained using a plurality of images of known objects, to classify the object encapsulated within the respective bounding polygon.Type: GrantFiled: November 29, 2016Date of Patent: May 14, 2019Assignee: SAP SEInventors: Waqas Ahmad Farooqi, Jonas Lipps, Eckehard Schmidt, Thomas Fricke, Nemrude Verzano
-
Patent number: 10282639Abstract: RGB-D data generated by at least one optical sensor for a field of view is received. Thereafter, the RGB-D data is bifurcated into (i) RGB data and (ii) depth data for the field of view. One or more bounding polygons are defined within the depth data that each characterize a window within the field of view encapsulating an object. The RGB data is then cropped using the bounding polygon(s). Image processing can later be applied to the cropped RGB data to identify at least one object therein. Related apparatus, systems, techniques and articles are also described.Type: GrantFiled: November 29, 2016Date of Patent: May 7, 2019Assignee: SAP SEInventors: Waqas Ahmad Farooqi, Jonas Lipps, Eckehard Schmidt, Thomas Fricke, Nemrude Verzano
-
Patent number: 10235594Abstract: Image color data for a field of view is received. Thereafter, color segmentation can be performed on the image color data to define at least one bounding polygon that minimizes an amount of free space within each bounding polygon. The at least one bounding polygon is then used to crop the image color data to result in cropped image color data. Image processing can then be applied to the cropped image color data to identify at least one object therein. Related apparatus, systems, techniques and articles are also described.Type: GrantFiled: November 29, 2016Date of Patent: March 19, 2019Assignee: SAP SEInventors: Waqas Ahmad Farooqi, Jonas Lipps, Eckehard Schmidt, Thomas Fricke, Nemrude Verzano
-
Patent number: 10032126Abstract: A system and method of purchasing goods and arranging delivery. The system receives, from a mobile device within a physical store, identifications, made by a customer using the mobile device, of at least two items in the physical store as items to be purchased. The items to be purchased include a first set of items to be delivered and a second set of items not to be delivered. The system receives, from the mobile device, a delivery address for a first set of items. The system receives payment for the first set of items and the second set of items in a single transaction. The system places items corresponding to the first set of items in a delivery pipeline for subsequent physical delivery to the delivery address and authorizes removal of the second set of items from the physical store.Type: GrantFiled: June 18, 2015Date of Patent: July 24, 2018Assignee: SAP SEInventors: Eckehard Schmidt, Filip Perisic
-
Publication number: 20180150727Abstract: RGB-D data generated by at least one optical sensor for a field of view is received. Thereafter, the RGB-D data is bifurcated into (i) RGB data and (ii) depth data for the field of view. One or more bounding polygons are defined within the depth data that each characterize a window within the field of view encapsulating an object. The RGB data is then cropped using the bounding polygon(s). Image processing can later be applied to the cropped RGB data to identify at least one object therein. Related apparatus, systems, techniques and articles are also described.Type: ApplicationFiled: November 29, 2016Publication date: May 31, 2018Inventors: Waqas Ahmad Farooqi, Jonas Lipps, Eckehard Schmidt, Thomas Fricke, Nemrude Verzano
-
Publication number: 20180150713Abstract: Combined color and depth data for a field of view is received. Thereafter, using at least one bounding polygon algorithm, at least one proposed bounding polygon is defined for the field of view. It can then be determined, using a binary classifier having at least one machine learning model trained using a plurality of images of known objects, whether each proposed bounding polygon encapsulates an object. The image data within each bounding polygon that is determined to encapsulate an object can then be provided to a first object classifier having at least one machine learning model trained using a plurality of images of known objects, to classify the object encapsulated within the respective bounding polygon.Type: ApplicationFiled: November 29, 2016Publication date: May 31, 2018Inventors: Waqas Ahmad Farooqi, Jonas Lipps, Eckehard Schmidt, Thomas Fricke, Nemrude Verzano
-
Publication number: 20180150716Abstract: Image color data for a field of view is received. Thereafter, color segmentation can be performed on the image color data to define at least one bounding polygon that minimizes an amount of free space within each bounding polygon. The at least one bounding polygon is then used to crop the image color data to result in cropped image color data. Image processing can then be applied to the cropped image color data to identify at least one object therein. Related apparatus, systems, techniques and articles are also described.Type: ApplicationFiled: November 29, 2016Publication date: May 31, 2018Inventors: Waqas Ahmad Farooqi, Jonas Lipps, Eckehard Schmidt, Thomas Fricke, Nemrude Verzano
-
Publication number: 20160371766Abstract: A system and method that integrate shopping online and in physical stores are provided. The system provides a virtual shopping cart for online shopping. The virtual shopping cart includes a listing of items selected from one or more online stores. The virtual shopping cart is accessible through a mobile device. The system activates a set of physical store related functions (sometimes referred to as “in-store functions”) when the mobile device enters the proximity of a physical store with devices configured to interface with the system. The system identifies a list of items in the virtual shopping cart that are available in the physical store. The system provides the list on the mobile device.Type: ApplicationFiled: June 18, 2015Publication date: December 22, 2016Inventors: Eckehard Schmidt, Filip Perisic
-
Publication number: 20160371650Abstract: A system and method of purchasing goods and arranging delivery. The system receives, from a mobile device within a physical store, identifications, made by a customer using the mobile device, of at least two items in the physical store as items to be purchased. The items to be purchased include a first set of items to be delivered and a second set of items not to be delivered. The system receives, from the mobile device, a delivery address for a first set of items. The system receives payment for the first set of items and the second set of items in a single transaction. The system places items corresponding to the first set of items in a delivery pipeline for subsequent physical delivery to the delivery address and authorizes removal of the second set of items from the physical store.Type: ApplicationFiled: June 18, 2015Publication date: December 22, 2016Inventors: Eckehard Schmidt, Filip Perisic
-
Publication number: 20160371606Abstract: A system and method of providing a virtual waiting line. The system determines that a mobile device associated with a particular account has entered a proximity of a physical store. The system assigns an initial position in a virtual waiting line to the particular account. The virtual waiting line includes an order of positions, from the front of the virtual waiting line to the back of the virtual waiting line. As accounts closer to the front of the virtual waiting line than the particular account leave the virtual waiting line, the system repeatedly moves the particular account to positions closer to the front of the virtual line. The system causes an electronic display of the mobile device to show a relationship of the position of the particular account to the front of the virtual waiting line.Type: ApplicationFiled: June 18, 2015Publication date: December 22, 2016Inventors: Eckehard Schmidt, Filip Perisic
-
Patent number: 4358331Abstract: A method of embedding semiconductor components in plastics comprising inserting components into the plastics material and thereafter hardening the plastics material by irradiation with high energy beams.Type: GrantFiled: March 5, 1980Date of Patent: November 9, 1982Assignee: Licentia Patent-Verwaltungs-G.m.b.H.Inventors: Eckehard Schmidt, Dieter Rusch, Manfred Tauber