Patent
Learning method and learning device for pooling ROI by using masking parameters to be used for mobile devices or compact networks via hardware optimization, and testing method and testing device using the same
Kim Kye-Hyeon,Kim Yongjoong,Kim Insu,Kim Hak-Kyoung,Nam Woonhyun,Boo Sukhoon,Sung Myungchul,Yeo Donghun,Ryu Wooju,Jang Taewoong,Jeong Kyungjoong,Je Hongmo,Cho Hojin +12 more
- 23 Jan 2019
4
TL;DR: In this article, a method for pooling at least one ROI by using one or more masking parameters is presented, which is applicable to mobile devices, compact networks, and the like via hardware optimization.
read more
Abstract: A method for pooling at least one ROI by using one or more masking parameters is provided. The method is applicable to mobile devices, compact networks, and the like via hardware optimization. The method includes steps of: (a) a computing device, if an input image is acquired, instructing a convolutional layer of a CNN to generate a feature map corresponding to the input image; (b) the computing device instructing an RPN of the CNN to determine the ROI corresponding to at least one object included in the input image by using the feature map; (c) the computing device instructing an ROI pooling layer of the CNN to apply each of pooling operations correspondingly to each of sub-regions in the ROI by referring to each of the masking parameters corresponding to each of the pooling operations, to thereby generate a masked pooled feature map.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Citations
Patent
Learning management device, learning management method, and imaging device
Haneda Kazuhiro,Ito Dai,Li Zhen,Osa Kazuhiko,Nonaka Osamu +4 more
- 05 Dec 2019
TL;DR: In this paper, a learning management device comprising an inference engine that is input with image data and acquires output for performing guidance display to a user using an inference model that has been specifically learned, and a processor that confirms if there is a specified relationship between a trial input image and trial guidance display for this input image under conditions that have been assumed at the time of use, and that determines whether or not it is necessary to relearn the inference model based on the result of this confirmation.
1
Patent
Video image ship detection method and system under influence of sea waves based on deep learning
Deng Lianbing
- 03 Jul 2020
TL;DR: In this paper, a video image ship detection method and system under the influence of sea waves based on deep learning is presented, which comprises the steps: marking video image data, including the marking of ship data and sea wave data; extracting an ROI (region of interest) to obtain position information of a ship in the image and a sea wave target easily confused with the ship; carrying out permutation and combination on every two ROIs of the region of interest, wherein the constructed double-branch convolutional neural network structure comprises an input layer, a plurality of hidden
1
Patent
Method and device for supporting administrators to processes of object detectors to provide logical driving
Kim Kye-Hyeon,Kim Yongjoong,Kim Hak-Kyoung,Nam Woonhyun,Boo Sukhoon,Sung Myungchul,Shin Dongsoo,Yeo Donghun,Ryu Wooju,Lee Myeong-Chun,Lee Hyungsoo,Jang Taewoong,Jeong Kyungjoong,Je Hongmo,Cho Hojin +14 more
- 14 Jul 2020
TL;DR: In this paper, a method for supporting at least one administrator to evaluate detecting processes of object detectors to provide logical grounds of an autonomous driving is provided, which includes steps of a computing device instructing convolutional layers, included in an object detecting CNN which has been trained before, to generate reference CNN feature maps.
Patent
Target detection method based on deep learning, and electronic apparatus
Wang Jianzong,Jia Xueli +1 more
- 07 Jan 2021
TL;DR: In this article, a target detection method based on deep learning is proposed, which comprises: acquiring a picture to be detected, inputting the picture into an improved VGG16 network to extract an image feature; inputting image feature into an ROI pooling network to perform pooling; and then classifying a target and a background by means of a fully connected layer network, so as to acquire category information and position information of the target.
References
Patent
System and Method for Extremely Efficient Image and Pattern Recognition and Artificial Intelligence Platform
Lotfi A. Zadeh,Saied Tadayon,Bijan Tadayon +2 more
- 12 Mar 2018
TL;DR: In this article, the first application of general-AI is described, which covers new algorithms, methods, and systems for: Artificial Intelligence; the first applications of General-AI. (versus Specific, Vertical, or Narrow-AI) (as humans can do) (which also includes Explainable-AI or XAI); addition of reasoning, inference, and cognitive layers/engines to learning module/engine/layer; soft computing; Information Principle; Stratification; Incremental Enlargement Principle; deep-level/detailed recognition, e.g.,
216
•Posted Content
End-to-End Saliency Mapping via Probability Distribution Prediction
TL;DR: In this paper, a new saliency map model is proposed which formulates a map as a generalized Bernoulli distribution and then trains a deep architecture to predict such maps using novel loss functions which pair the softmax activation function with measures designed to compute distances between probability distributions.
134
Patent
Sparse map for autonomous vehicle navigation
Amnon Shashua,Yoram Gdalyahu,Ofer Springer,Aran Reisman,Daniel Braunstein +4 more
- 10 Feb 2016
TL;DR: In this article, a sparse map for autonomous vehicle navigation along a road segment is presented, which includes a polynomial representation of a target trajectory for the autonomous vehicle along the road segment and a plurality of predetermined landmarks associated with the road segments.
105
Patent
Subcategory-aware convolutional neural networks for object detection
Wongun Choi,Yuanqing Lin,Yu Xiang,Silvio Savarese +3 more
- 03 Nov 2016
TL;DR: In this paper, a computer-implemented method for detecting objects by using subcategory-aware convolutional neural networks (CNNs) is presented, which includes generating object region proposals from an image by a region proposal network (RPN) which utilizes subcategory information, and classifying and refining the object regions proposals by an object detection network (ODN) that simultaneously performs object category classification, subcategory classification, and bounding box regression.
104
Patent
Systems and methods for end-to-end object detection
Lichao Huang,Yi Yang,Yafeng Deng,Yinan Yu +3 more
- 05 Aug 2016
TL;DR: In this article, a unified end-to-end detection pipeline for object detection is presented, which achieves impressive performance in detecting very small and highly overlapped objects in face and car images.
96