公开(公告)号：US20190287297A1

公开(公告)日：2019-09-19

申请号：US16273618

申请日：2019-02-12

申请人： Recogni Inc.

发明人： Shabarivas Abhiram ,  Gilles J. C. A. Backhus ,  Eugene M. Feinberg ,  Berend Ozceri ,  Martin Stefan Patz

IPC分类号： G06T17/05 ,  G06T7/593 ,  H04N7/18 ,  G06N20/00

摘要： Systems, methods, and machine-readable media for determining a three-dimensional environment model of the environment of one or more camera devices, in which image processing for inferring the model may be performed at the camera devices, are described.

公开(公告)号：US11915126B2

公开(公告)日：2024-02-27

申请号：US16948164

申请日：2020-09-04

申请人： Recogni Inc.

发明人： Jian hui Huang ,  James Michael Bodwin ,  Pradeep R. Joginipally ,  Shabarivas Abhiram ,  Gary S. Goldman ,  Martin Stefan Patz ,  Eugene M. Feinberg ,  Berend Ozceri

IPC分类号： G06N3/04 ,  G06N3/06 ,  G06N3/063 ,  G06F7/50 ,  G06F7/544 ,  G06N3/0464

CPC分类号： G06N3/063 ,  G06F7/50 ,  G06F7/5443 ,  G06N3/0464

摘要： Dynamic data quantization may be applied to minimize the power consumption of a system that implements a convolutional neural network (CNN). Under such a quantization scheme, a quantized representation of a 3×3 array of m-bit activation values may include 9 n-bit mantissa values and one exponent shared between the n-bit mantissa values (n

公开(公告)号：US20240143988A1

公开(公告)日：2024-05-02

申请号：US18410736

申请日：2024-01-11

申请人： Recogni Inc.

发明人： Jian hui Huang ,  James Michael Bodwin ,  Pradeep R. Joginipally ,  Shabarivas Abhiram ,  Gary S. Goldman ,  Martin Stefan Patz ,  Eugene M. Feinberg ,  Berend Ozceri

IPC分类号： G06N3/063 ,  G06F7/50 ,  G06F7/544 ,  G06N3/0464

CPC分类号： G06N3/063 ,  G06F7/50 ,  G06F7/5443 ,  G06N3/0464

摘要： Dynamic data quantization may be applied to minimize the power consumption of a system that implements a convolutional neural network (CNN). Under such a quantization scheme, a quantized representation of a 3×3 array of m-bit activation values may include 9 n-bit mantissa values and one exponent shared between the n-bit mantissa values (n

公开(公告)号：US10740964B2

公开(公告)日：2020-08-11

申请号：US16273618

申请日：2019-02-12

申请人： Recogni Inc.

发明人： Shabarivas Abhiram ,  Gilles J. C. A. Backhus ,  Eugene M. Feinberg ,  Berend Ozceri ,  Martin Stefan Patz

IPC分类号： G06T17/05 ,  G06N20/00 ,  H04N7/18 ,  G06T7/593 ,  G06T7/246 ,  G06T7/579

摘要： A three-dimensional model of the environment of one or more camera devices is determined, in which image processing for inferring the model may be performed at the one or more camera devices.

公开(公告)号：US20220076104A1

公开(公告)日：2022-03-10

申请号：US16948164

申请日：2020-09-04

申请人： Recogni Inc.

发明人： Jian hui Huang ,  James Michael Bodwin ,  Pradeep R. Joginipally ,  Shabarivas Abhiram ,  Gary S. Goldman ,  Martin Stefan Patz ,  Eugene M. Feinberg ,  Berend Ozceri

IPC分类号： G06N3/063 ,  G06F7/544 ,  G06F7/50

摘要： Dynamic data quantization may be applied to minimize the power consumption of a system that implements a convolutional neural network (CNN). Under such a quantization scheme, a quantized representation of a 3×3 array of m-bit activation values may include 9 n-bit mantissa values and one exponent shared between the n-bit mantissa values (n