International Business Machines Corporation

Samuel Mathew Jawaharlal

Sathya Santhar

Balamurugaramanathan Sivaramalingam

Sarbajit Kumar Rakshit

Sridevi Kannan

Arunachalam Narayanan

Mukundan Sundararajan

Bala Priyadarshini Muralidharan

Suvedhahari Velusamy

Kothagorla Lakshmana Rao

Santosh Shanmukh

A computer-implemented method for multimedia modification is disclosed. The computer-implemented method includes classifying one or more objects detected within a user's field of view through an augmented reality environment. The computer-implemented method further includes determining a context of the user based, at least in part, on the one or more classified objects detected within the user's field of view. The computer-implemented method further includes generating a priority score for the one or more classified objects based, at least in part, on the context of the user. The computer-implemented method further includes modifying an object detected within the user's field of view based, at least in part, on the priority score of the object.

Contextual priority based multimedia modification

According to one embodiment, a method, computer system, and computer program product for digital image enhancement is provided. The embodiment may include receiving a digital image being navigated by a user. The embodiment may include identifying an image content navigation pattern of the user and a context of the digital image. Based on the image content navigation pattern and context, the embodiment may include identifying one or more objects of the digital image on which the user will perform a predicted zoom-in action. The embodiment may include identifying one or more distributed computing resources for digital image enhancement of the identified one or more objects. The embodiment may include performing digital image enhancement of the identified one or more objects. Upon the user performing a zoom-in action on an identified object which matches a predicted zoom-in action, the embodiment may include displaying a digitally enhanced image of the identified object.

Generative adversarial network based digital image enhancement through distributed resources

An embodiment for transmitting and regenerating compressed media is provided. The embodiment may include receiving an image selected by a sender to be transmitted. The embodiment may also include sending the image to a pre-trained generative adversarial network (GAN) model. The embodiment may further include identifying a first set of one or more pixels in the image assigned to one or more high priority Class of Service (CoS) groups and a second set of one or more pixels in the image assigned to one or more low priority CoS groups. The embodiment may also include compressing the one or more pixels in the second set. The embodiment may further include transmitting the one or more pixels in the first set and the compressed one or more pixels in the second set to a recipient in an order from a highest priority CoS group to a lowest priority CoS group.

Compressed media transmission and high quality regeneration

Deviations in a 3D printing process can be detected during printing via sensors or by user observation. A user can manipulate and evaluate any deviations from a 3D model via a simulation of the work-in-progress object via virtual reality. The user can also make changes to the model, resulting in the printer completing the same work-in-progress object based on the revised model.

3D printing with virtual reality integration

A method, computer program, and computer system are provided for text summarization that maintains emotional content. Data corresponding to text to be summarized and a target emotion to be maintained in the text to be summarized is received. The target emotion is encoded as an emotion probability vector. One or more words that correspond to the target emotion that is encoded in the emotion probability vector are identified from a dictionary. A text summary to be associated with the text to be summarized is generated based on the one or more identified words.

Text summarization with emotion conditioning

A computer-implemented method according to one embodiment includes causing an environment generator of a Generative Adversarial Network (GAN) to generate realistic training environments, and causing a first discriminator of the GAN to determine whether the realistic training environments are real or fake. In response to a determination that an accuracy of the first discriminator at determining whether the realistic training environments are real or fake is within a predetermined range, the environment generator is caused to generate a first realistic environment. The method further includes causing the first realistic environment to be shared with an agent of a reinforcement learning (RL) algorithm and a second discriminator, and receiving, from the agent of the RL algorithm and the second discriminator, feedback associated with the first realistic environment. The environment generator is caused to generate a second realistic environment based on the feedback associated with the first realistic environment.

Training an environment generator of a generative adversarial network (GAN) to generate realistic environments that incorporate reinforcement learning (RL) algorithm feedback

An approach for generating an optimized video of a speaker, translated from a source language into a target language with the speaker's lips synchronized to the translated speech, while balancing optimization of the translation into a target language. A source video may be fed into a neural machine translation model. The model may synthesize a plurality of potential translations. The translations may be received by a generative adversarial network which generates video for each translation and classifies the translations as in-sync or out of sync. A lip-syncing score may be for each of the generated videos that are classified as in-sync.

Optimization of lip syncing in natural language translated video

A computer-implemented method, according to one embodiment, includes: determining whether a predetermined version of a source script is available. In response to determining that a predetermined version of the source script is available, it is used to condition a first processor, and instructions are sent to the conditioned first processor to generate a translated copy of the source script by translating the words in the source script from a source language to a target language. Instructions are also sent to a second processor to determine a distribution of metrics associated with the speech of each of the actors in the source audio file. The distributions are used to condition a third processor, and instructions are sent to the conditioned third processor to generate an audio file that includes words spoken in the target language. Furthermore, instructions are sent to merge the generated audio file with the video file.

Generating audio files based on user generated scripts and voice components

Provided is a system and method for moving a virtual object within virtual space in response to an external input supplied by a user. A machine learning model may predict a movement of the virtual object and implement such movement in a next frame of the virtual space. An example operation may include one or more of receiving a measurement of an external input of a user with respect to a virtual object displayed in virtual space, predicting, via execution of a machine learning model, a movement of the virtual object in the virtual space in response to the external input of the user based on the measurement of the external input of the user, and moving the virtual object in the virtual space based on the predicted movement of the virtual object by the machine learning model.

Creating a virtual object response to a user input

A video file is detected by a computer system. The video file that is to be provided to one or more client devices. The video file contains a video stream that includes a plurality of video images. A first video image of the plurality of video images is reconstructed based on a first machine learning technique. The first machine learning technique is based on one or more video images that occur temporally before the first video image in the video stream. A reconstruction status of the first video image of the plurality of video images is identified based on the video file and based on a second machine learning technique. An altered video file is generated in response to the reconstruction status and based on the video file.

Video size reduction by reconstruction

Image data encryption by receiving first image data corresponding to a first image having a first image size, compressing the first image data, yielding second image data corresponding to a second image having a second image size, augmenting the second image data yielding third image data corresponding to a third image having the first image size, determining coordinates of a location of the second image within the third image, encrypting the third image data according to the coordinates, providing the encrypted third image data to a decoder by a first communications channel, and providing the coordinates of the second image within the third image to the decoder by a second communications channel.

Image encryption

A system, method, and computer program product for implementing digital script modification is provided. The method includes generating image sequences associated with textual content of a digital story. Multiple contextual dimensions are identified within the textual content and a group of dimensions are selected. The image sequences in combination with the group of dimensions are expanding or contracted and image sequences are altered based on detected interactions with the group of dimensions. Dimensions are extracted from the group of dimensions during presentation of the digital story and a scriptwriter is enabled to modify the dimensions. The image sequences are modified and a hardware interface device is enabled to interact with various image sequences and alter the multiple contextual dimensions. The textual content of the digital story is dynamically altered.

Generative adversarial network implemented digital script modification

A processor may initiate a recording. The processor may segment the recording into one or more segments. The processor may determine, based on the identification of a primary object in a first segment of the recording, a first bit rate for the first segment of the first recording. The processor may preload one or more subsequent segments that include the primary object at the first bit rate. The processor may preload each of the one or more subsequent segments with a secondary object at a second bit rate. The second bit rate may be lower than the first bit rate. The processor may display the recording to the user.

Partial loading of media based on context

An example operation may include one or more of receiving, by a blockchain node, a first transaction comprising an asset identifier and an expiration date, verifying the expiration date has not changed from one or more previous transactions comprising the asset identifier, calculating an expected delivery date for an asset corresponding to the asset identifier, validating the first transaction in response to the expected delivery date not later than the expiration date. and committing the first transaction to a blockchain.

Perishable asset tracking for blockchain

Dynamic checkpointing can include determining, using a processor, a process criticality corresponding to a process configured to execute on computer hardware. The process criticality indicates a priority of the process relative to at least one other process configured to execute on the computer hardware. A checkpoint schedule can be generated, using the processor and based on the process criticality, for checkpointing the process when it executes on the computer hardware.

Growing community of inventors

Chennai, India

Samuel Mathew Jawaharlal