International Business Machines Corporation

Christophe Menichetti

Ivan Deleuze

Guilhaume Garcia

Geoffrey Pascal

Pascal Vezolle

Olivier Boehler

Nicolas Tallet

Cyril Armand

Using a trained image classification model, a first image captured using an image capture device is classified, the trained image classification model executing in the same system as the image capture device, the classifying producing an inappropriate content classification of the first image. Using a trained feature extraction model, a pattern within a portion of the first image is recognized from the first image, the trained feature extraction model executing in the same system as the image capture device, the pattern predefined as inappropriate content. Based on the classification and the pattern, the first image is adjusted.

Content based on-device image adjustment

A method, computer system, and a computer program product for shine visualization is provided. The present invention may include recognizing an object in a digital image loaded on a user device. The present invention may include assigning a shine index to the recognized object. The present invention may include determining, based on a plurality of pixel values corresponding to the recognized object, a direction of light relative to the recognized object. The present invention may include tracking a position of a user's eyes viewing the digital image on the user device. The present invention may include in response to detecting a movement in the position of the user's eyes, applying, in real-time, at least one filter to the recognized object to simulate a shining effect of the recognized object in the digital image.

Shine visualization

A method to automatically position entered content into a matching entry field of an online form accessed from a computing device. A position of an active cursor relative to entry fields of the online form is determined. An expected character and format set is determined for the entry fields of the online form. A first character and format set entered at a first entry field is analyzed to determine whether the entered character and format set matches an expected characters and format set, based on machine learning training for entry field content and format. In response to the character and format set entered in the first entry field failing to match the expected character and format set for the first entry field, entering the first character and format set in a second entry field with a highest match probability between the expected and entered character and format set.

Alignment of content entry with online form fields

An operation of an action zone in an image displayed on a display device is received. The operation has a defined location in the image and a current time for an operation of the action zone. A time delay is applied to the current time to obtain a selection time for the operation of the action zone, by calculating the selection time as the current time minus the time delay. A map with a time stamp immediately prior to the selection time is determined from maps of the image. Each map defines, for a specific time stamp, a location for each action zone in the image. An action zone defined in the determined map and matching the defined location of the operation of the action zone in the image is operated.

Display of images with action zones

A method and system. An image, displayed on a display device, includes action zones. Maps of the image are stored. Each map defines, for a specific time stamp, a location for each action zone in the image. An operation of an action zone in the image displayed on the display device is received. The operation has a defined location in the image and a current time for an operation of the action zone. A time delay is applied to the current time to obtain a selection time for the operation of the action zone, by calculating the selection time as the current time minus the time delay. A map with a time stamp immediately prior to the selection time is determined from the maps. An action zone defined in the determined map and matching the defined location of the operation of the action zone in the image is operated.

Growing community of inventors

Montpellier, France

Christophe Menichetti