Invention Grant
- Patent Title: Automated selection of priors for training of detection convolutional neural networks
-
Application No.: US16772358Application Date: 2018-02-05
-
Publication No.: US11657283B2Publication Date: 2023-05-23
- Inventor: Konstantin Rodyushkin , Alexander Bovyrin
- Applicant: INTEL CORPORATION
- Applicant Address: US CA Santa Clara
- Assignee: INTEL CORPORATION
- Current Assignee: INTEL CORPORATION
- Current Assignee Address: US CA Santa Clara
- Agency: Hanley, Flight & Zimmerman, LLC
- International Application: PCT/IB2018/000280 2018.02.05
- International Announcement: WO2019/150158A 2019.08.08
- Date entered country: 2020-06-12
- Main IPC: G06N3/08
- IPC: G06N3/08 ; G06K9/62 ; G06V10/75 ; G06N3/082

Abstract:
An example apparatus for selecting priors includes a training set receiver to receive a training dataset. The apparatus includes a prior generator to generate a set of redundant priors based on the training dataset. The apparatus includes an intermediate trainer to train a detection CNN using the set of redundant priors. The apparatus includes a score and location receiver to send all training samples of the training dataset to the trained detection CNN and receive responses for all of the redundant priors in the set of redundant priors. The apparatus includes a subset selector to select a subset of the set of redundant priors based on the responses.
Public/Granted literature
- US20210081797A1 AUTOMATED SELECTION OF PRIORS FOR TRAINING OF DETECTION CONVOLUTIONAL NEURAL NETWORKS Public/Granted day:2021-03-18
Information query