A Kernel Correlation-Based Approach to Adaptively Acquire Local Features for Learning 3D Point Clouds

Song, Yupeng; He, Fazhi; Duan, Yun‐Bo; Liang, Yunsheng; Yan, Xiaohu

doi:10.1016/j.cad.2022.103196

Cited by 51 publications

(5 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In future research, considering potential co-editing applications, [55][56][57][58][59] we plan to explore the other collaboration capabilities of address space transformation, such as group undo, 46 and extend the consistency algorithms to 2D/3D collaborative systems. [60][61][62][63][64][65][66][67][68]…”

Section: Discussionmentioning

confidence: 99%

Self‐compressing object sequence for consistency maintenance in co‐editors

Cai

Yang

et al. 2022

Softw Pract Exp

Self Cite

View full text Add to dashboard Cite

Co-editors are a class of human-centered collaborative systems that allow multiple geographically dispersed people to freely and concurrently edit shared documents at the same time over networks. The most important building co-editors technique is consistency maintenance. Even though OT (operational transformation) is a widely adopted consistency maintenance technique, researchers and practitioners have persistently explored alternative techniques to OT. One of the representative techniques is AST (Address Space Transformation), which replaces OT's transformation components with basic manipulation on an additional object sequence. However, the additional object sequence causes an unbounded metadata overhead problem. To solve the problem, this work proposes an optimized AST algorithm called ASTO, which is equipped with a specially designed self-compressing object sequence. ASTO can automatically compress the object sequence and keep the metadata overhead under control.The effectiveness and feasibility of ASTO are verified by simulation experiments and a publicly accessible prototype system.

show abstract

Section: Discussionmentioning

confidence: 99%

Self‐compressing object sequence for consistency maintenance in co‐editors

Cai

Yang

et al. 2022

Softw Pract Exp

Self Cite

View full text Add to dashboard Cite

show abstract

“…In future studies, we will focus on adversarial defense methods that are robust to adaptive attacks. In addition, we will extend our hybrid adversarial training method to 3D engineering applications [36][37][38].…”

Section: Discussionmentioning

confidence: 99%

A hybrid adversarial training for deep learning model and denoising network resistant to adversarial examples

Ryu

Choi

2022

Appl Intell

View full text Add to dashboard Cite

Deep neural networks (DNNs) are vulnerable to adversarial attacks that generate adversarial examples by adding small perturbations to the clean images. To combat adversarial attacks, the two main defense methods used are denoising and adversarial training. However, both methods result in the DNN having lower classification accuracy for clean images than conventionally trained DNN models. To overcome this problem, we propose a hybrid adversarial training (HAT) method that trains the denoising network and DNN model simultaneously. The proposed HAT method uses both clean images and adversarial examples denoised by the denoising network and non-denoised clean images and adversarial examples to train the DNN model. The results of experiments conducted on the MNIST, CIFAR-10, CIFAR-100, and GTSRB datasets show that the HAT method results in a higher classification accuracy than both conventional training with a denoising network and previous adversarial training methods. They also indicate that training with the HAT method results in average improvements in robustness of 0.84%, 27.33%, 28.99%, and 17.61% against adversarial attacks compared with several state-of-the-art adversarial training methods on the MNIST, CIFAR-10, CIFAR-100, and GTSRB datasets, respectively. Thus, the proposed HAT method results in improved robustness for DNNs against a wider range of adversarial attacks.

show abstract

“…In recent years, with the development of deep learning [7][8][9][10], the 3D reconstruction method combined with deep learning has gained attention. A typical method is to directly output 3D models from images by training a neural network [11][12][13][14][15]. Another popular method recently is to learn a neural network which approximates the implicit function [16][17][18][19][20][21].…”

Section: Related Workmentioning

confidence: 99%

3D reconstruction based on hierarchical reinforcement learning with transferability

Fan

et al. 2023

ICA

Self Cite

View full text Add to dashboard Cite

3D reconstruction is extremely important in CAD (computer-aided design)/CAE (computer-aided Engineering)/CAM (computer-aided manufacturing). For interpretability, reinforcement learning (RL) is used to reconstruct 3D shapes from images by a series of editing actions. However, typical applications of RL for 3D reconstruction face problems. The search space will increase exponentially with the action space due to the curse of dimensionality, which leads to low performance, especially for complex action spaces in 3D reconstruction. Additionally, most works involve training a specific agent for each shape class without learning related experiences from others. Therefore, we present a hierarchical RL approach with transferability to reconstruct 3D shapes (HRLT3D). First, actions are grouped into macro actions that can be chosen by the top-agent. Second, the task is accordingly decomposed into hierarchically simplified sub-tasks solved by sub-agents. Different from classical hierarchical RL (HRL), we propose a sub-agent based on augmented state space (ASS-Sub-Agent) to replace a set of sub-agents, which can speed up the training process due to shared learning and having fewer parameters. Furthermore, the ASS-Sub-Agent is more easily transferred to data of other classes due to the augmented diverse states and the simplified tasks. The experimental results on typical public dataset show that the proposed HRLT3D performs overwhelmingly better than recent baselines. More impressingly, the experiments also demonstrate the extreme transferability of our approach among data of different classes.

show abstract

A Kernel Correlation-Based Approach to Adaptively Acquire Local Features for Learning 3D Point Clouds

Cited by 51 publications

References 25 publications

Self‐compressing object sequence for consistency maintenance in co‐editors

Self‐compressing object sequence for consistency maintenance in co‐editors

A hybrid adversarial training for deep learning model and denoising network resistant to adversarial examples

3D reconstruction based on hierarchical reinforcement learning with transferability

Contact Info

Product

Resources

About