Resumes
Resumes
Bhavin Nayak Phones & Addresses
-
Mountain View, CA
-
San Jose, CA
-
Santa Clara, CA
-
Tempe, AZ
Publications
Us Patents
Partially-Frozen Neural Networks For Efficient Computer Vision Systems
View pageUS Patent:
20200293870, Sep 17, 2020
Filed:
May 28, 2020
Appl. No.:
16/886103
Inventors:
- Santa Clara CA, US
Bhavin V. Nayak - San Jose CA, US
Joao Peralta Moreira - Mountain View CA, US
Gilad Michael - Sunnyvale CA, US
Bhavin V. Nayak - San Jose CA, US
Joao Peralta Moreira - Mountain View CA, US
Gilad Michael - Sunnyvale CA, US
Assignee:
Intel Corporation - Santa Clara CA
International Classification:
G06N 3/063
G06N 3/04
G06N 3/08
G06K 9/62
G06N 3/04
G06N 3/08
G06K 9/62
Abstract:
An apparatus to facilitate partially-frozen neural networks for efficient computer vision systems is disclosed. The apparatus includes a frozen core to store fixed weights of a machine learning model, one or more trainable cores coupled to the frozen core, the one or more trainable cores comprising multipliers for trainable weights of the machine learning model, and wherein the alpha blending layer includes a trainable alpha blending parameter, and wherein the trainable alpha blending parameter is a function of a trainable parameter, a sigmoid function, and outputs of frozen and trainable blocks in a preceding layer of the machine learning model.
Attenuating Visual Artifacts Of Image Processing Systems Using Adversarial Networks On-The-Fly
View pageUS Patent:
20200226429, Jul 16, 2020
Filed:
Mar 27, 2020
Appl. No.:
16/833263
Inventors:
- Santa Clara CA, US
Gilad Michael - Sunnyvale CA, US
Joao Peralta Moreira - Mountain View CA, US
Bhavin Nayak - Santa Clara CA, US
Gilad Michael - Sunnyvale CA, US
Joao Peralta Moreira - Mountain View CA, US
Bhavin Nayak - Santa Clara CA, US
International Classification:
G06K 9/62
G06N 3/08
G06N 3/04
G06N 3/08
G06N 3/04
Abstract:
An apparatus, method, and a computer readable medium for attenuating visual artifacts in processed images. An annotated dataset of images to be processed by an image processing system is created. An adversarial control network is trained to operate as an image quality expert in classifying images. After the adversarial control network has been trained, the adversarial control network is used to supervise the image processing system on-the-fly.
Processing Images Using Hybrid Infinite Impulse Response (Iir) And Finite Impulse Response (Fir) Convolution Block
View pageUS Patent:
20200074691, Mar 5, 2020
Filed:
Nov 5, 2019
Appl. No.:
16/674512
Inventors:
- Santa Clara CA, US
Sushma Rao - Sunnyvale CA, US
Avi Kalderon - Holon, IL
Bhavin Nayak - Santa Clara CA, US
Joao Peralta Moreira - Montain View CA, US
Pavel Kounitsky - Holon, IL
Ben Berlin - Kfar Saba, IL
Gilad Michael - Sunnyvale CA, US
Sushma Rao - Sunnyvale CA, US
Avi Kalderon - Holon, IL
Bhavin Nayak - Santa Clara CA, US
Joao Peralta Moreira - Montain View CA, US
Pavel Kounitsky - Holon, IL
Ben Berlin - Kfar Saba, IL
Gilad Michael - Sunnyvale CA, US
Assignee:
INTEL CORPORATION - Santa Clara CA
International Classification:
G06T 9/00
G06T 3/40
G06T 1/20
G06T 5/00
G06T 5/20
G06T 3/40
G06T 1/20
G06T 5/00
G06T 5/20
Abstract:
An example apparatus for processing images includes a hybrid infinite impulse response—finite impulse response (IIR-FIR) convolution block to receive an image and generate processed image information. The hybrid IIR-FIR convolution block includes a vertical infinite impulse response (IIR) component to approximate a vertical convolution when processing the image.