US 12,223,412 B2
Dynamic matrix convolution with channel fusion
Yinpeng Chen, Sammamish, WA (US); Xiyang Dai, Seattle, WA (US); Mengchen Liu, Redmond, WA (US); Dongdong Chen, Bellevue, WA (US); Lu Yuan, Redmond, WA (US); Zicheng Liu, Bellevue, WA (US); Ye Yu, Redmond, WA (US); Mei Chen, Bellevue, WA (US); and Yunsheng Li, San Diego, CA (US)
Assigned to Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed by Microsoft Technology Licensing, LLC, Redmond, WA (US)
Filed on Dec. 16, 2020, as Appl. No. 17/123,697.
Prior Publication US 2022/0188595 A1, Jun. 16, 2022
Int. Cl. G06N 3/04 (2023.01); G06F 17/16 (2006.01); G06V 10/70 (2022.01)
CPC G06N 3/04 (2013.01) [G06F 17/16 (2013.01); G06V 10/70 (2022.01)] 20 Claims
OG exemplary drawing
 
1. A computer system for automatic feature detection, comprising:
a processor;
a communication interface; and
a memory configured to hold instructions executable by the processor to:
instantiate a dynamic convolution neural network;
receive input data via the communication interface;
execute the dynamic convolution neural network to:
compress the input data from an input space having a dimensionality equal to a predetermined number of channels into an intermediate space having a dimensionality less than the number of channels;
dynamically fuse the channels into an intermediate representation within the intermediate space; and
expand the intermediate representation from the intermediate space to an expanded representation in an output space having a higher dimensionality than the dimensionality of the intermediate space; and
automatically detect features in the input data based on the expanded representation in the output space.