Fast raycasting using a compound deep image for VPL range determination

The perception and identification of visual stimuli from the environment is a fundamental capacity of autonomous mobile robots. Current deep learning techniques make it possible to identify and segment objects of interest in an image. This paper presents a novel algorithm to segment the object’s space from a deep segmentation of an image taken by a 3D camera. The proposed approach solves the boundary pixel problem that appears when a direct mapping from segmented pixels to their correspondence in the point cloud is used. We validate our approach by comparing baseline approaches using real images taken by a 3D camera, showing that our method outperforms their results in terms of accuracy and reliability. As an application of the proposed algorithm, we present a semantic mapping approach for a mobile robot’s indoor environments.

Download Full-text

Using Deep Image Priors to Generate Counterfactual Explanations

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP) ◽

10.1109/icassp39728.2021.9413636 ◽

2021 ◽

Author(s):

Vivek Narayanaswamy ◽

Jayaraman J. Thiagarajan ◽

Andreas Spanias

Keyword(s):

Deep Image ◽

Image Priors

Download Full-text

DIDarknet: A Contemporary Approach to Detect and Characterize the Darknet Traffic using Deep Image Learning

2020 the 10th International Conference on Communication and Network Security ◽

10.1145/3442520.3442521 ◽

2020 ◽

Cited By ~ 1

Author(s):

Arash Habibi Lashkari ◽

Gurdip Kaur ◽

Abir Rahali

Keyword(s):

Deep Image ◽

Contemporary Approach ◽

Image Learning

Download Full-text

Real-Time Adversarial Attack Detection with Deep Image Prior Initialized as a High-Level Representation Based Blurring Network

Electronics ◽

10.3390/electronics10010052 ◽

2020 ◽

Vol 10 (1) ◽

pp. 52

Author(s):

Richard Evan Sutanto ◽

Sukho Lee

Keyword(s):

Neural Network ◽

Attack Detection ◽

Detection Methods ◽

Defense System ◽

Image Prior ◽

The Neural Network ◽

Adversarial Examples ◽

Deep Image ◽

Adversarial Attack ◽

High Level

Several recent studies have shown that artificial intelligence (AI) systems can malfunction due to intentionally manipulated data coming through normal channels. Such kinds of manipulated data are called adversarial examples. Adversarial examples can pose a major threat to an AI-led society when an attacker uses them as means to attack an AI system, which is called an adversarial attack. Therefore, major IT companies such as Google are now studying ways to build AI systems which are robust against adversarial attacks by developing effective defense methods. However, one of the reasons why it is difficult to establish an effective defense system is due to the fact that it is difficult to know in advance what kind of adversarial attack method the opponent is using. Therefore, in this paper, we propose a method to detect the adversarial noise without knowledge of the kind of adversarial noise used by the attacker. For this end, we propose a blurring network that is trained only with normal images and also use it as an initial condition of the Deep Image Prior (DIP) network. This is in contrast to other neural network based detection methods, which require the use of many adversarial noisy images for the training of the neural network. Experimental results indicate the validity of the proposed method.

Download Full-text

DISC: Deep Image Saliency Computing via Progressive Representation Learning

IEEE Transactions on Neural Networks and Learning Systems ◽

10.1109/tnnls.2015.2506664 ◽

2016 ◽

Vol 27 (6) ◽

pp. 1135-1149 ◽

Cited By ~ 80

Author(s):

Tianshui Chen ◽

Liang Lin ◽

Lingbo Liu ◽

Xiaonan Luo ◽

Xuelong Li

Keyword(s):

Representation Learning ◽

Image Saliency ◽

Deep Image

Download Full-text

A Draft Genome Sequence of Pseudomonas syringae pv. tomato T1 Reveals a Type III Effector Repertoire Significantly Divergent from That of Pseudomonas syringae pv. tomato DC3000

Molecular Plant-Microbe Interactions ◽

10.1094/mpmi-22-1-0052 ◽

2009 ◽

Vol 22 (1) ◽

pp. 52-62 ◽

Cited By ~ 101

Author(s):

Nalvo F. Almeida ◽

Shuangchun Yan ◽

Magdalen Lindeberg ◽

David J. Studholme ◽

David J. Schneider ◽

...

Keyword(s):

Host Range ◽

Host Specificity ◽

Pseudomonas Syringae ◽

Genome Sequence ◽

Draft Genome ◽

Draft Genome Sequence ◽

Tomato Dc3000 ◽

Type Iii ◽

Host Range Determination ◽

Range Determination

Diverse gene products including phytotoxins, pathogen-associated molecular patterns, and type III secreted effectors influence interactions between Pseudomonas syringae strains and plants, with additional yet uncharacterized factors likely contributing as well. Of particular interest are those interactions governing pathogen-host specificity. Comparative genomics of closely related pathogens with different host specificity represents an excellent approach for identification of genes contributing to host-range determination. A draft genome sequence of Pseudomonas syringae pv. tomato T1, which is pathogenic on tomato but nonpathogenic on Arabidopsis thaliana, was obtained for this purpose and compared with the genome of the closely related A. thaliana and tomato model pathogen P. syringae pv. tomato DC3000. Although the overall genetic content of each of the two genomes appears to be highly similar, the repertoire of effectors was found to diverge significantly. Several P. syringae pv. tomato T1 effectors absent from strain DC3000 were confirmed to be translocated into plants, with the well-studied effector AvrRpt2 representing a likely candidate for host-range determination. However, the presence of avrRpt2 was not found sufficient to explain A. thaliana resistance to P. syringae pv. tomato T1, suggesting that other effectors and possibly type III secretion system–independent factors also play a role in this interaction.

Download Full-text