-
公开(公告)号：US20240361985A1
公开(公告)日：2024-10-31
申请号：US18603806
申请日：2024-03-13
发明人： Jinnan DING , Xiangtao WU
CPC分类号： G06F7/5318 , G06F7/50 , H03K19/21
摘要： Disclosed are an operation method of multiplier, an electronic device, and a storage medium. The method includes: determining a plurality of input data sets of the multiplier and an encoding manner for the multiplier; determining at least one low-order input data set in the plurality of input data sets; determining a carry compensation term corresponding to the at least one low-order input data set based on the at least one low-order input data set and the encoding manner; determining a target partial product array based on the carry compensation term corresponding to the at least one low-order input data set and the plurality of input data sets; and determining a product operation result for each input data set based on the target partial product array. According to this disclosure, multiplication operations with multiple precision may be implemented by using one multiplier, thereby reducing hardware resource consumption and hardware area.
-
公开(公告)号：US20240303040A1
公开(公告)日：2024-09-12
申请号：US18595690
申请日：2024-03-05
发明人： Yibo HE , Lei XIAO , Honghe TAN
CPC分类号： G06F7/5443 , G06F9/30134
摘要： Disclosed are a method for processing neural network feature map using a plurality of accelerators. The method includes: reading first feature data about the neural network feature map from first shift register array in first accelerator among a plurality of neural network accelerators, and first weight data corresponding to the first feature data from first buffer; performing preset operation on the first feature data and first weight data using the first accelerator, to obtain a first operation result; shifting, according to preset shift rule, first overlapping feature data in the first feature data and required by a second accelerator to a second shift register array of the second accelerator; and performing a preset operation on the second feature data from the second shift register array including the first overlapping feature data and the read second weight data using the second accelerator, to obtain a second operation result.
-
3.发明申请公开(公告)号：US20220189059A1
公开(公告)日：2022-06-16
申请号：US17540624
申请日：2021-12-02
发明人： Wei Sui , Qian Zhang , Hongmei Zhu
摘要： Embodiments of the present disclosure disclose an image-based pose determination method and apparatus, a computer readable storage medium, and an electronic device. The method include: acquiring a current image frame captured by a camera disposed on a moving object and a historical image frame captured before the current image frame; determining a first relative camera pose between the current image frame and the historical image frame; determining a virtual binocular image based on the first relative camera pose; and determining current pose information of the camera based on preset visual odometry and the virtual binocular image. According to the embodiments of the present disclosure, the virtual binocular image may be generated based on a monocular image, thus achieving effects of binocular visual odometry without using a binocular camera, thereby reducing costs. Moreover, monocular odometry may be enabled to obtain real physical scale of a space where the moving object is located, thereby improving accuracy of the monocular odometry in determining a position and a pose of the camera.
-
4.发明申请公开(公告)号：US20220182072A1
公开(公告)日：2022-06-09
申请号：US17526335
申请日：2021-11-15
发明人： Zhenjiang Wang , Jianjun Li , Zhuoran Zhao , Chang Huang
IPC分类号： H03M7/30
摘要： Embodiments of the present disclosure disclose a data compression method and apparatus, a computer-readable storage medium, and an electronic device. The method includes: converting each data in a to-be-compressed data set into binary data in a preset format; determining a to-be-compressed bit and a significant bit for the each data in the to-be-compressed data set based on a sequence of all bits of the binary data; determining a compression bit width corresponding to the to-be-compressed data set based on bit widths of the significant bits; compressing the each data in the to-be-compressed data set based on the compression bit width, to obtain a compressed data set; and generating attribute information of the compressed data set. According to the embodiments of the present disclosure, the significant bit can be determined based on the sequence of the all bits without adjusting orders of the bits of the binary data. In this way, a data compression process is simplified, and efficiency of data compression is improved.
-
公开(公告)号：US10936487B2
公开(公告)日：2021-03-02
申请号：US15919207
申请日：2018-03-12
发明人： Delin Li , Zhenjiang Wang , Wenhui Cao , Kun Lin , Liang Chen , Jianjun Li , Chang Huang
IPC分类号： G06F12/06
摘要： A method and apparatus are disclosed to perform the circular addressing to emulate a virtually unlimited memory space despite the fixed capacity of a physical memory by readdressing the portion of the data that exceeds the pre-defined length of the circular addressing region to another pre-defined address in the circular addressing region. Data segments in a data sample can be loaded and computed with recalculated circular addresses for different applications.
-
公开(公告)号：US20190278707A1
公开(公告)日：2019-09-12
申请号：US15919207
申请日：2018-03-12
发明人： Delin Li , Zhenjiang Wang , Wenhui Cao , Kun Lin , Liang Chen , Jianjun Li , Chang Huang
IPC分类号： G06F12/06
摘要： A method and apparatus are disclosed to perform the circular addressing to emulate a virtually unlimited memory space despite the fixed capacity of a physical memory by readdressing the portion of the data that exceeds the pre-defined length of the circular addressing region to another pre-defined address in the circular addressing region. Data segments in a data sample can be loaded and computed with recalculated circular addresses for different applications.
-
7.发明公开公开(公告)号：US20240354427A1
公开(公告)日：2024-10-24
申请号：US18637802
申请日：2024-04-17
发明人： Shunli CHENG , Jianjun LI , Jing LI
CPC分类号： G06F21/604 , G06F21/6209
摘要： Disclosed are a data register access method, a computer readable storage medium and electronic device. The method includes: determining a first virtual machine identifier and first address of a to-be-accessed data register from a register access request generated by a target virtual machine; determining, based on the first virtual machine identifier, a target data register group and target protection register from at least two data register groups and a protection register group; determining an access permission of the target virtual machine to the target data register group using the target protection register; and determining, based on the first address, the to-be-accessed data register from the target data register group in response to that the access permission is access allowed, and accessing the to-be-accessed data register using the target virtual machine, whereby isolation between register groups is implemented on hardware, improving efficiency of accessing the register by the virtual machine.
-
8.发明公开公开(公告)号：US20240321268A1
公开(公告)日：2024-09-26
申请号：US18589639
申请日：2024-02-28
发明人： Yichen GONG , Ziyang WANG
CPC分类号： G10L15/1815 , G06F3/013 , G10L15/063 , G10L15/22 , G06V40/18 , G10L2015/0635
摘要： Embodiments of the present disclosure disclose a human-machine interaction method, an apparatus, a computer-readable storage medium, and an electronic device, wherein the method includes: in response to receiving target interaction information, performing semantic recognition on the target interaction information to obtain target semantic information; in response to the target semantic information being incomplete, determining whether there is to-be-combined semantic information cached in a preset semantic state record library; in response to presence of the to-be-combined semantic information in the semantic state record library, generating complete semantic information based on the target semantic information and the to-be-combined semantic information; updating the to-be-combined semantic information in the semantic state record library based on the target semantic information; and based on the complete semantic information, determining a target controlled object and a control mode for controlling the target controlled object, and generating a control instruction corresponding to the control mode.
-
公开(公告)号：US20240220387A1
公开(公告)日：2024-07-04
申请号：US18462975
申请日：2023-09-07
发明人： Wenxing LI , Jianjun LI , Qingyu LIU , Jing LI
CPC分类号： G06F11/3072 , G06F11/079
摘要： Disclosed are a fault processing method and apparatus, an electronic device and storage medium. The method includes: classifying at least one piece of to-be-processed fault information to determine fault classes to which all the to-be-processed fault information respectively belongs, wherein the to-be-processed fault information includes a fault identifier of a to-be-processed fault; arbitrating the to-be-processed fault information included at all fault classes respectively, to determine target to-be-processed fault information respectively corresponding to all the fault classes; and performing corresponding fault processing based on all the target to-be-processed fault information. According to the embodiments of the present disclosure, faults at a plurality of fault classes can be processed concurrently, avoiding occurrence of an interrupt storm or a situation of no responses, thereby effectively improving fault processing efficiency, and resolving problems such as interrupt storm or starving to death due to lack of responses that are easily caused in related technologies.
-
10.发明授权公开(公告)号：US12026910B2
公开(公告)日：2024-07-02
申请号：US17540624
申请日：2021-12-02
发明人： Wei Sui , Qian Zhang , Hongmei Zhu
CPC分类号： G06T7/74 , G06T7/50 , G06T2207/20084 , G06T2207/30244
摘要： Embodiments of the present disclosure disclose an image-based pose determination method and apparatus, a computer readable storage medium, and an electronic device. The method include: acquiring a current image frame captured by a camera disposed on a moving object and a historical image frame captured before the current image frame; determining a first relative camera pose between the current image frame and the historical image frame; determining a virtual binocular image based on the first relative camera pose; and determining current pose information of the camera based on preset visual odometry and the virtual binocular image. According to the embodiments of the present disclosure, the virtual binocular image may be generated based on a monocular image, thus achieving effects of binocular visual odometry without using a binocular camera, thereby reducing costs. Moreover, monocular odometry may be enabled to obtain real physical scale of a space where the moving object is located, thereby improving accuracy of the monocular odometry in determining a position and a pose of the camera.
-
-
-
-
-
-
-
-
-
搜索结果
23 条记录 (耗时 0.02 秒)
