Zobrazeno 1 - 10
of 55
pro vyhledávání: '"Tieyong Cao"'
Publikováno v:
Alexandria Engineering Journal, Vol 108, Iss , Pp 50-59 (2024)
In object segmentation, the existence of hard-classified-pixels limits the segmentation performance. Focusing on these hard pixels through assigning different weights to pixel loss can guide the learning of segmentation model effectively. Existing lo
Externí odkaz:
https://doaj.org/article/cc4deb9c423a43d8aa5435d3f37b2e99
Publikováno v:
International Journal of Computational Intelligence Systems, Vol 17, Iss 1, Pp 1-16 (2024)
Abstract 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 adversaria
Externí odkaz:
https://doaj.org/article/19206d3a309a44ea8e8093627c459d08
Publikováno v:
Complex & Intelligent Systems, Vol 10, Iss 5, Pp 6825-6837 (2024)
Abstract Significant structural differences in DNN-based object detectors hinders the transferability of adversarial attacks. Studies show that intermediate features extracted by the detector contain more model-independent information, and disrupting
Externí odkaz:
https://doaj.org/article/372584eec4da4ccdafc23686cbfd546c
Publikováno v:
Complex & Intelligent Systems, Vol 10, Iss 5, Pp 6545-6557 (2024)
Abstract Self-distillation method guides the model learning via transferring knowledge of the model itself, which has shown the advantages in object segmentation. However, it has been proved that uncertain pixels with predicted probability close to 0
Externí odkaz:
https://doaj.org/article/a1657d44e824414e91a6412015c964b8
Publikováno v:
Journal of King Saud University: Computer and Information Sciences, Vol 36, Iss 1, Pp 101901- (2024)
Deep neural networks (DNNs) are vulnerable to adversarial attacks, which can cause security risks in computer information systems. Feature disruption attacks, as a typical form of adversarial attack, optimize adversarial examples by disrupting the in
Externí odkaz:
https://doaj.org/article/ee5d1a332b0c486fb70b2303e70c0df0
Publikováno v:
IET Computer Vision, Vol 17, Iss 3, Pp 341-351 (2023)
Abstract Most self‐distillation methods need complex auxiliary teacher structures and require lots of training samples in object segmentation task. To solve this challenging, a self‐distillation object segmentation method via frequency domain kno
Externí odkaz:
https://doaj.org/article/b23897b99ad84c72aa01a290710db13b
Publikováno v:
PeerJ Computer Science, Vol 9, p e1435 (2023)
Self-distillation methods utilize Kullback-Leibler divergence (KL) loss to transfer the knowledge from the network itself, which can improve the model performance without increasing computational resources and complexity. However, when applied to sal
Externí odkaz:
https://doaj.org/article/b7fd831394c643c0a1a0b37993eb28ec
Publikováno v:
IET Computer Vision, Vol 16, Iss 3, Pp 205-217 (2022)
Abstract In skeleton‐based action recognition, the graph convolutional network (GCN) has achieved great success. Modelling skeleton data in a suitable spatial‐temporal way and designing the adjacency matrix are crucial aspects for GCN‐based met
Externí odkaz:
https://doaj.org/article/266c6d4fbd704ca2b6cc1adf8a7a7031
Publikováno v:
IEEE Access, Vol 7, Pp 54321-54329 (2019)
In recent years, neural network-based voice conversion methods have been rapidly developed, and many different models and neural networks have been applied in parallel voice conversion. However, the over-smoothing of parametric methods [e.g., bidirec
Externí odkaz:
https://doaj.org/article/45b145412b6d4c1c832e1127d8f80b2a
Publikováno v:
IEEE Access, Vol 6, Pp 71455-71463 (2018)
Non-acoustic sensors are widely used in speech signal processing tasks, and their immunity to the background acoustic noise shows great benefits to traditional speech enhancement. To avoid using acoustic speech disturbed by strong noise, spectra mapp
Externí odkaz:
https://doaj.org/article/6baa622d1e9744a9a11079955e0c197f