-
公开(公告)号：US20160206197A1
公开(公告)日：2016-07-21
申请号：US15000032
申请日：2016-01-19
Inventor： Shantanu Sinha , Hyunsung Park , Albert Redo-Sanchez , Matthew Everett Lawson , Nickolaos Savidis , Pushyami Rachapudi , Ramesh Raskar , Vincent Patalano, II
CPC classification number: A61B3/117 , A61B3/0025 , A61B3/1025 , A61B3/14 , G02B5/10
Abstract: A projector and one or more optical components project a light pattern that scans at least a portion of an anterior segment of an eye of a user, while one or more cameras capture images of the anterior segment. During each scan, different pixels in the projector emit light at different times, causing the light pattern to repeatedly change orientation relative to the eye and thus to illuminate multiple different cross-sections of the anterior segment. The cameras capture images of each cross-section from a total of at least two different vantage points relative to the head of the user. The position of the projector, optical components and cameras relative to the head of the user remains substantially constant throughout each entire scan.
Abstract translation: 投影仪和一个或多个光学部件投射扫描用户眼睛的前部的至少一部分的光图案，而一个或多个相机拍摄前部片段的图像。 在每次扫描期间，投影仪中的不同像素在不同时间发光，导致光图案相对于眼睛重复地改变取向，从而照亮前段的多个不同横截面。 相机相对于用户的头部，相机从总共至少两个不同的有利位置拍摄每个横截面的图像。 投影仪，光学部件和相机相对于用户头部的位置在整个扫描期间保持基本上恒定。
-
公开(公告)号：US11669737B2
公开(公告)日：2023-06-06
申请号：US16934685
申请日：2020-07-21
Applicant: Massachusetts Institute of Technology
Inventor： Otkrist Gupta , Ramesh Raskar
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.
-
公开(公告)号：US11481635B2
公开(公告)日：2022-10-25
申请号：US16862494
申请日：2020-04-29
Applicant: Massachusetts Institute of Technology , Otkrist Gupta
Inventor： Praneeth Vepakomma , Abhishek Singh , Otkrist Gupta , Ramesh Raskar
IPC: G06N3/08
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.
-
公开(公告)号：US20200349443A1
公开(公告)日：2020-11-05
申请号：US16862494
申请日：2020-04-29
Applicant: Otkrist Gupta , Massachusetts Institute of Technology
Inventor： Praneeth Vepakomma , Abhishek Singh , Otkrist Gupta , Ramesh Raskar
IPC: G06N3/08
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.
-
公开(公告)号：US10561309B2
公开(公告)日：2020-02-18
申请号：US15849559
申请日：2017-12-20
Applicant: Massachusetts Institute of Technology
Inventor： Anshuman Das , Ramesh Raskar
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.
-
Patent Agency Ranking
公开(公告)号：US10557705B2
公开(公告)日：2020-02-11
申请号：US16278769
申请日：2019-02-19
Applicant: Massachusetts Institute of Technology
Inventor： Achuta Kadambi , Vage Taamazyan , Boxin Shi , Ramesh Raskar
Abstract: A 3D imaging system uses a depth sensor to produce a coarse depth map, and then uses the coarse depth map as a constraint in order to correct ambiguous surface normals computed from polarization cues. The imaging system outputs an enhanced depth map that has a greater depth resolution than the coarse depth map. The enhanced depth map is also much more accurate than could be obtained from the depth sensor alone. In many cases, the imaging system extracts the polarization cues from three polarized images. Thus, in many implementations, the system takes only three extra images—in addition to data used to generate the coarse depth map—in order to dramatically enhance the coarse depth map.
-
27.发明授权公开(公告)号：US10337993B2
公开(公告)日：2019-07-02
申请号：US15487435
申请日：2017-04-14
Applicant: Massachusetts Institute of Technology
Inventor： Ayush Bhandari , Christopher Barsi , Achuta Kadambi , Ramesh Raskar
Abstract: A light source may illuminate a scene with amplitude-modulated light. The scene may include fluorescent material. The amplitude modulation may be periodic, and the frequency of the amplitude modulation may be swept. During the sweep, a time-of-flight sensor may take measurements of light returning from the scene. A computer may calculate, for each pixel in the sensor, a vector of complex numbers. Each complex number in the vector may encode phase and amplitude of light incident at the pixel and may correspond to measurements taken at a given frequency in the sweep. A computer may, based on phase of the complex numbers for a pixel, calculate fluorescence lifetime and scene depth of a scene point that corresponds to the pixel.
-
公开(公告)号：US20190187788A1
公开(公告)日：2019-06-20
申请号：US16278100
申请日：2019-02-17
Applicant: Massachusetts Institute of Technology
Inventor： Tristan Swedish , Karin Roesch , Ramesh Raskar
CPC classification number: G06F3/013 , G06F3/005 , G06K9/00604
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.
-
公开(公告)号：US20170372201A1
公开(公告)日：2017-12-28
申请号：US15630944
申请日：2017-06-22
Applicant: Massachusetts Institute of Technology
Inventor： Otkrist Gupta , Ramesh Raskar
CPC classification number: G06N3/08 , G06N3/0454 , G06N3/084 , G06N20/00
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.
-
30.发明授权公开(公告)号：US11016309B2
公开(公告)日：2021-05-25
申请号：US16506762
申请日：2019-07-09
Applicant: Massachusetts Institute of Technology
IPC: G02B27/58 , G01N21/3586 , G02B27/10 , G01B9/02 , G01N21/47
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.
-
-
-
-
-
-
-
-
-
Search Results
52
records
(took 0.036 seconds)
