Patents by Inventor Ramesh Raskar

Ramesh Raskar 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).

  • Patent number: 11481635
    Abstract: A distributed deep learning network may prevent an attacker from reconstructing raw data from activation outputs of an intermediate layer of the network. To achieve this, the loss function of the network may tend to reduce distance correlation between raw data and the activation outputs. For instance, the loss function may be the sum of two terms, where the first term is weighted distance correlation between raw data and activation outputs of a split layer of the network, and the second term is weighted categorical cross entropy of actual labels and label predictions. Distance correlation with the entire raw data may be minimized. Alternatively, distance correlation with only with certain features of the raw data may be minimized, in order to ensure attribute-level privacy. In some cases, a client computer calculates decorrelated representations of raw data before sharing information about the data with external computers.
    Type: Grant
    Filed: April 29, 2020
    Date of Patent: October 25, 2022
    Assignee: Massachusetts Institute of Technology
    Inventors: Praneeth Vepakomma, Abhishek Singh, Otkrist Gupta, Ramesh Raskar
  • Publication number: 20220198673
    Abstract: A computer-implemented method for computing a prediction on images of a scene includes: receiving one or more polarization raw frames of a scene, the polarization raw frames being captured with a polarizing filter at a different linear polarization angle; extracting one or more first tensors in one or more polarization representation spaces from the polarization raw frames; and computing a prediction regarding one or more optically challenging objects in the scene based on the one or more first tensors in the one or more polarization representation spaces.
    Type: Application
    Filed: March 8, 2022
    Publication date: June 23, 2022
    Inventors: Agastya KALRA, Vage TAAMAZYAN, Supreeth Krishna RAO, Kartik VENKATARAMAN, Ramesh RASKAR, Achuta KADAMBI
  • Patent number: 11302012
    Abstract: A computer-implemented method for computing a prediction on images of a scene includes: receiving one or more polarization raw frames of a scene, the polarization raw frames being captured with a polarizing filter at a different linear polarization angle; extracting one or more first tensors in one or more polarization representation spaces from the polarization raw frames; and computing a prediction regarding one or more optically challenging objects in the scene based on the one or more first tensors in the one or more polarization representation spaces.
    Type: Grant
    Filed: August 28, 2020
    Date of Patent: April 12, 2022
    Assignee: BOSTON POLARIMETRICS, INC.
    Inventors: Agastya Kalra, Vage Taamazyan, Supreeth Krishna Rao, Kartik Venkataraman, Ramesh Raskar, Achuta Kadambi
  • Patent number: 11181623
    Abstract: A time-of-flight imaging system may output light with a modulation frequency in the gigahertz band, to illuminate a range target. This high-frequency illumination may enable extremely precise—e.g., micron-scale—depth measurements. The system may modulate reflected light from the range target, to create a beat tone that has a frequency in the hertz band. In some cases, the modulated light in the gigahertz band is created by a first modulator and the beat tone in the hertz band is created by a second modulator. In some cases, the modulated light in the gigahertz band is created by an upshift cascade of modulators and the beat tone in the hertz band is created by a downshift cascade of modulators. A photodetector may measure the low-frequency beat tone. From this beat tone, phase of the signal and depth of the range target may be extracted.
    Type: Grant
    Filed: September 30, 2018
    Date of Patent: November 23, 2021
    Assignee: Massachusetts Institute of Technology
    Inventors: Achuta Kadambi, Tomohiro Maeda, Ayush Bhandari, Barmak Heshmat Dehkordi, Ramesh Raskar
  • Publication number: 20210356572
    Abstract: A multi-modal sensor system includes: an underlying sensor system; a polarization camera system configured to capture polarization raw frames corresponding to a plurality of different polarization states; and a processing system including a processor and memory, the processing system being configured to control the underlying sensor system and the polarization camera system, the memory storing instructions that, when executed by the processor, cause the processor to: control the underlying sensor system to perform sensing on a scene and the polarization camera system to capture a plurality of polarization raw frames of the scene; extract first tensors in polarization representation spaces based on the plurality of polarization raw frames; and compute a characterization output based on an output of the underlying sensor system and the first tensors in polarization representation spaces.
    Type: Application
    Filed: October 7, 2020
    Publication date: November 18, 2021
    Inventors: Achuta KADAMBI, Ramesh RASKAR, Kartik VENKATARAMAN, Supreeth Krishna RAO, Agastya KALRA
  • Patent number: 11156828
    Abstract: One embodiment of the present application sets forth a wearable device that includes a display comprising a plurality of pixels and configured to emit light, and a micro-lens array located adjacent to the display, and configured to produce a light field by altering the light emitted by the display, where at least one of the display or the micro-lens array is configured to move from a first position to a second position that aligns a first pixel in the plurality of pixels relative to the micro-lens array.
    Type: Grant
    Filed: July 5, 2018
    Date of Patent: October 26, 2021
    Assignee: FACEBOOK TECHNOLOGIES, LLC
    Inventors: Ahmad Byagowi, Ramesh Raskar, Andrew Hamilton Coward, Ethan Kroll Gordon
  • Publication number: 20210264607
    Abstract: A computer-implemented method for computing a prediction on images of a scene includes: receiving one or more polarization raw frames of a scene, the polarization raw frames being captured with a polarizing filter at a different linear polarization angle; extracting one or more first tensors in one or more polarization representation spaces from the polarization raw frames; and computing a prediction regarding one or more optically challenging objects in the scene based on the one or more first tensors in the one or more polarization representation spaces.
    Type: Application
    Filed: August 28, 2020
    Publication date: August 26, 2021
    Inventors: Agastya KALRA, Vage TAAMAZYAN, Supreeth Krishna RAO, Kartik VENKATARAMAN, Ramesh RASKAR, Achuta KADAMBI
  • Patent number: 11061977
    Abstract: In one embodiment, a method includes receiving data associated with the user from a content rendering device associated with a user of an online social network, where the content rendering device receives signals from a broadcast service provider system, identifying personalized content items that are of interest to the user based at least on the received data associated with the user, sending the identified personalized content items to the content rendering device, where the identified personalized content items are used by the content rendering device to determine which content items to cache, ranking content objects in the online social network based at least on the identified personalized content items, sending one or more content objects that have higher ranks than the other content objects to the broadcast service provider system, where the one or more content objects are sent to the content rendering device by the broadcast service provider system.
    Type: Grant
    Filed: December 28, 2018
    Date of Patent: July 13, 2021
    Assignee: Facebook, Inc.
    Inventor: Ramesh Raskar
  • Patent number: 11016309
    Abstract: A sample may be illuminated in such a way that light passes through the sample, reflects from a set of reflectors, passes through the sample again and travels to a light sensor. The reflectors may be staggered in depth beneath the sample, each reflector being at a different depth. Light reflecting from each reflector, respectively, may arrive at the light sensor during a different time interval than that in which light reflecting from other reflectors arrives—or may have a different phase than that of light reflecting from the other reflectors. The light sensor may separately measure light reflecting from each reflector, respectively. The reflectors may be extremely small, and the separate reflections from the different reflectors may be combined in a super-resolved image. The super-resolved image may have a spatial resolution that is better than that indicated by the diffraction limit.
    Type: Grant
    Filed: July 9, 2019
    Date of Patent: May 25, 2021
    Assignee: Massachusetts Institute of Technology
    Inventors: Barmak Heshmat Dehkordi, Albert Redo-Sanchez, Gordon Moseley Andrews, Ramesh Raskar
  • Patent number: 10918272
    Abstract: An otoscope may project a temporal sequence of phase-shifted fringe patterns onto an eardrum, while a camera in the otoscope captures images. A computer may calculate a global component of these images. Based on this global component, the computer may output an image of the middle ear and eardrum. This image may show middle ear structures, such as the stapes and incus. Thus, the otoscope may “see through” the eardrum to visualize the middle ear. The otoscope may project another temporal sequence of phase-shifted fringe patterns onto the eardrum, while the camera captures additional images. The computer may subtract a fraction of the global component from each of these additional images. Based on the resulting direct-component images, the computer may calculate a 3D map of the eardrum.
    Type: Grant
    Filed: December 31, 2019
    Date of Patent: February 16, 2021
    Assignee: Massachusetts Institute of Technology
    Inventors: Anshuman Das, Ramesh Raskar
  • Patent number: 10845874
    Abstract: In one embodiment, a computing system may determine that a performance metric of an eye tracking system is below a performance threshold. The eye tracking system may be associated with a head-mounted display worn by a user. In response to the determination that the performance metric is below the performance threshold, the system may identify one or more contents being displayed by the head-mounted display. The system may access one or more properties associated with the one or more contents. The system may predict a vergence distance of the user based at least on the one or more properties associated with the one or more display contents. The system may adjust one or more configurations based on the predicted vergence distance of the user.
    Type: Grant
    Filed: December 5, 2019
    Date of Patent: November 24, 2020
    Assignee: Facebook Technologies, LLC
    Inventors: Ramesh Raskar, Neeraj Choubey
  • Publication number: 20200351233
    Abstract: In one embodiment, a method includes determining if notifications to be sent to user would benefit from being delivered by haptic stimulation under a current context. This determination may be made by accessing historical notification data of how the user previously responded to notifications in a similar context, and ranking conversion scores for each of one or more haptic-enabled delivery channels, wherein a conversion score indicates a probability of the user interacting with the notification. The most appropriate haptic message-delivery channel is selected based on the scores and historical data, and the notification is sent accordingly.
    Type: Application
    Filed: July 16, 2020
    Publication date: November 5, 2020
    Inventors: Ramesh Raskar, Nafissa Yakubova, Ahmad Byagowi, Marie K. Herring
  • Publication number: 20200349435
    Abstract: A deep neural network may be trained on the data of one or more entities, also know as Alices. An outside computing entity, also known as a Bob, may assist in these computations, without receiving access to Alices' data. Data privacy may be preserved by employing a “split” neural network. The network may comprise an Alice part and a Bob part. The Alice part may comprise at least three neural layers, and the Bob part may comprise at least two neural layers. When training on data of an Alice, that Alice may input her data into the Alice part, perform forward propagation though the Alice part, and then pass output activations for the final layer of the Alice part to Bob. Bob may then forward propagate through the Bob part. Similarly, backpropagation may proceed backwards through the Bob part, and then through the Alice part of the network.
    Type: Application
    Filed: July 21, 2020
    Publication date: November 5, 2020
    Inventors: Otkrist Gupta, Ramesh Raskar
  • Publication number: 20200349443
    Abstract: A distributed deep learning network may prevent an attacker from reconstructing raw data from activation outputs of an intermediate layer of the network. To achieve this, the loss function of the network may tend to reduce distance correlation between raw data and the activation outputs. For instance, the loss function may be the sum of two terms, where the first term is weighted distance correlation between raw data and activation outputs of a split layer of the network, and the second term is weighted categorical cross entropy of actual labels and label predictions. Distance correlation with the entire raw data may be minimized. Alternatively, distance correlation with only with certain features of the raw data may be minimized, in order to ensure attribute-level privacy. In some cases, a client computer calculates decorrelated representations of raw data before sharing information about the data with external computers.
    Type: Application
    Filed: April 29, 2020
    Publication date: November 5, 2020
    Inventors: Praneeth Vepakomma, Abhishek Singh, Otkrist Gupta, Ramesh Raskar
  • Patent number: 10802583
    Abstract: A video camera captures images of retroreflection from the retina of an eye. These images are captured while the eye rotates. Thus, different images are captured in different rotational positions of the eye. A computer calculates, for each image, the eye's direction of gaze. In turn, the direction of gaze is used to calculate the precise location of a small region of the retina at which the retroflection occurs. A computer calculates a digital image of a portion of the retina by summing data from multiple retroreflection images. The digital image of the retina may be used for many practical applications, including medical diagnosis and biometric identification. In some scenarios, the video camera captures detailed images of the retina of a subject, while the subject is so far away that the rest of the subject's face is below the diffraction limit of the camera.
    Type: Grant
    Filed: February 17, 2019
    Date of Patent: October 13, 2020
    Assignee: Massachusetts Institute of Technology
    Inventors: Tristan Swedish, Karin Roesch, Ramesh Raskar
  • Patent number: 10796190
    Abstract: A sensor may measure light reflecting from a multi-layered object at different times. A digital time-domain signal may encode the measurements. Peaks in the signal may be identified. Each identified peak may correspond to a layer in the object. For each identified peak, a short time window may be selected, such that the time window includes a time at which the identified peak occurs. A discrete Fourier transform of that window of the signal may be computed. A frequency frame may be computed for each frequency in a set of frequencies in the transform. Kurtosis for each frequency frame may be computed. A set of high kurtosis frequency frames may be averaged, on a pixel-by-pixel basis, to produce a frequency image. Text characters that are printed on a layer of the object may be recognized in the frequency image, even though the layer is occluded.
    Type: Grant
    Filed: August 9, 2018
    Date of Patent: October 6, 2020
    Assignees: Massachusetts Institute of Technology, Georgia Tech Research Corporation
    Inventors: Barmak Heshmat Dehkordi, Albert Redo-Sanchez, Ramesh Raskar, Alireza Aghasi, Justin Romberg
  • Patent number: 10755172
    Abstract: A deep neural network may be trained on the data of one or more entities, also know as Alices. An outside computing entity, also known as a Bob, may assist in these computations, without receiving access to Alices' data. Data privacy may be preserved by employing a “split” neural network. The network may comprise an Alice part and a Bob part. The Alice part may comprise at least three neural layers, and the Bob part may comprise at least two neural layers. When training on data of an Alice, that Alice may input her data into the Alice part, perform forward propagation though the Alice part, and then pass output activations for the final layer of the Alice part to Bob. Bob may then forward propagate through the Bob part. Similarly, backpropagation may proceed backwards through the Bob part, and then through the Alice part of the network.
    Type: Grant
    Filed: June 22, 2017
    Date of Patent: August 25, 2020
    Assignee: Massachusetts Institute of Technology
    Inventors: Otkrist Gupta, Ramesh Raskar
  • Patent number: 10752158
    Abstract: A pulsed laser may illuminate a scene that is obscured by dense, dynamic and heterogeneous fog. Light may reflect back to a time-resolved camera. Each pixel of the camera may detect a single photon during each frame. The imaging system may accurately determine reflectance and depth of the fog-obscured target, without any calibration or prior knowledge of the scene depth. The imaging system may perform a probabilistic algorithm that exploits the fact that times of arrival of photons reflected from fog have a Gamma distribution that is different than the Gaussian distribution of times of arrival of photons reflected from the target. The probabilistic algorithm may take into account times of arrival of all types of measured photons, including scattered and un-scattered photons.
    Type: Grant
    Filed: February 7, 2019
    Date of Patent: August 25, 2020
    Assignee: Massachusetts Institute of Technology
    Inventors: Guy Satat, Matthew Tancik, Ramesh Raskar
  • Patent number: 10743234
    Abstract: In one embodiment, a method includes receiving, from a sender node associated with a mesh network, a request to send a message to one or more recipient nodes, the wireless mesh network comprising a plurality of nodes, detecting a triggering condition associated with the wireless mesh network, predicting a routing path from the sender node to each of the one or more recipient nodes via the wireless mesh network through one or more relay nodes of the plurality of nodes based on proximity information and network information associated with the mesh network, and sending the message to the one or more recipient nodes via the one or more relay nodes of the wireless mesh network.
    Type: Grant
    Filed: January 5, 2018
    Date of Patent: August 11, 2020
    Assignee: Facebook, Inc.
    Inventors: Sai Sri Sathya, Ramesh Raskar, Mayank Raj, Pritesh Sankhe
  • Patent number: 10742585
    Abstract: In one embodiment, a method includes determining if notifications to be sent to user would benefit from being delivered by haptic stimulation under a current context. This determination may be made by accessing historical notification data of how the user previously responded to notifications in a similar context, and ranking conversion scores for each of one or more haptic-enabled delivery channels, wherein a conversion score indicates a probability of the user interacting with the notification. The most appropriate haptic message-delivery channel is selected based on the scores and historical data, and the notification is sent accordingly.
    Type: Grant
    Filed: January 5, 2018
    Date of Patent: August 11, 2020
    Assignee: Facebook, Inc.
    Inventors: Ramesh Raskar, Nafissa Yakubova, Ahmad Byagowi, Marie K. Herring