Results 51 to 60 of about 26,718 (222)

Image Denoising via CNNs: An Adversarial Approach

, 2017
Is it possible to recover an image from its noisy version using convolutional neural networks? This is an interesting problem as convolutional layers are generally used as feature detectors for tasks like classification, segmentation and object detection.
Babu, R. Venkatesh, Divakar, Nithish
core   +1 more source

Patch-wise++ Perturbation for Adversarial Targeted Attacks

CoRR, 2020
Although great progress has been made on adversarial attacks for deep neural networks (DNNs), their transferability is still unsatisfactory, especially for targeted attacks. There are two problems behind that have been long overlooked: 1) the conventional setting of $T$ iterations with the step size of $ε/T$ to comply with the $ε$-constraint.
Lianli Gao, Qilong Zhang, Jingkuan Song, Heng Tao Shen   +3 more
openaire   +2 more sources

Domain‐Aware Implicit Network for Arbitrary‐Scale Remote Sensing Image Super‐Resolution

Advanced Intelligent Discovery, EarlyView.
Although existing arbitrary‐scale image super‐resolution methods are flexible to reconstruct images with arbitrary scales, the characteristic of training distribution is neglected that there exists domain shift between samples of various scales. In this work, a Domain‐Aware Implicit Network (DAIN) is proposed to handle it from the perspective of domain
Xiaoxuan Ren, Qian Jiang, Xin Jin, Puming Wang, Xing Chu, Shaowen Yao, Shengfa Miao   +6 more
wiley   +1 more source

Enhancing the Transferability of Adversarial Patch via Alternating Minimization

International Journal of Computational Intelligence Systems
Adversarial patches, a type of adversarial example, pose serious security threats to deep neural networks (DNNs) by inducing erroneous outputs. Existing gradient stabilization methods aim to stabilize the optimization direction of adversarial examples ...
Yang Wang, Lei Chen, Zhen Yang, Tieyong Cao   +3 more
doaj   +1 more source

Deep Learning‐Assisted Coherent Raman Scattering Microscopy

Advanced Intelligent Discovery, EarlyView.
The analytical capabilities of coherent Raman scattering microscopy are augmented through deep learning integration. This synergistic paradigm improves fundamental performance via denoising, deconvolution, and hyperspectral unmixing. Concurrently, it enhances downstream image analysis including subcellular localization, virtual staining, and clinical ...
Jianlin Liu, Rujia Liu, Xinyan Zhang, Xun Chen, Shuhua Yue   +4 more
wiley   +1 more source

Infrared Adversarial Patch Generation Based on Reinforcement Learning

Mathematics
Recently, there has been an increasing concern about the vulnerability of infrared object detectors to adversarial attacks, where the object detector can be easily spoofed by adversarial samples with aggressive patches.
Shuangju Zhou, Yang Li, Wenyi Tan, Chenxing Zhao, Xin Zhou, Quan Pan   +5 more
doaj   +1 more source

High-Quality Facial Photo-Sketch Synthesis Using Multi-Adversarial Networks

, 2018
Synthesizing face sketches from real photos and its inverse have many applications. However, photo/sketch synthesis remains a challenging problem due to the fact that photo and sketch have different characteristics. In this work, we consider this task as
Patel, Vishal M., Sindagi, Vishwanath A., Wang, Lidan   +2 more
core   +1 more source

DPatch: An Adversarial Patch Attack on Object Detectors [PDF]

, 2018
Object detectors have emerged as an indispensable module in modern computer vision systems. In this work, we propose DPatch -- a black-box adversarial-patch-based attack towards mainstream object detectors (i.e. Faster R-CNN and YOLO). Unlike the original adversarial patch that only manipulates image-level classifier, our DPatch simultaneously attacks ...
Xin Liu 0075, Huanrui Yang, Ziwei Liu 0002, Linghao Song, Yiran Chen 0001, Hai Li 0001   +5 more
openaire   +2 more sources

Detecting Patch Adversarial Attacks with Image Residuals

CoRR, 2020
We introduce an adversarial sample detection algorithm based on image residuals, specifically designed to guard against patch-based attacks. The image residual is obtained as the difference between an input image and a denoised version of it, and a discriminator is trained to distinguish between clean and adversarial samples.
Marius Arvinte, Ahmed H. Tewfik, Sriram Vishwanath   +2 more
openaire   +2 more sources

Real‐Time Multicolor Fluorescence Microscopy via Cross‐Channel Acquisition and Deep‐Learning‐Based Inference

Advanced Intelligent Discovery, EarlyView.
Sequential multicolor fluorescence imaging in dynamic microsystems is constrained by acquisition speed and excitation dose. This study introduces a real‐time framework to reconstruct spectrally separated channels from reduced cross‐channel acquisitions (frames containing mixed spectral contributions).
Juan J. Huaroto, Yuxin Jin, Nicholas Roy, Sarthak Misra   +3 more
wiley   +1 more source