检索结果-内蒙古大学图书馆

Task-Oriented 6-DoF Grasp Pose Detection in Clutters

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Wang, An-Lan Chen, Nuo Lin, Kun-Yu Li, Yuan-Ming Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China

In general, humans would grasp an object differently for different tasks, e.g., "grasping the handle of a knife to cut" vs. "grasping the blade to hand over". In the field of robotic grasp pose detection research, some existing works consider this task-oriented grasping and made some progress, but they are generally constrained by low-DoF gripper type or non-cluttered setting, which is not applicable for human assistance in real life. With an aim to get more general and practical grasp models, in this paper, we investigate the problem named Task-Oriented 6-DoF Grasp Pose Detection in Clutters (TO6DGC), which extends the task-oriented problem to a more general 6-DOF Grasp Pose Detection in Cluttered (multi-object) scenario. To this end, we construct a large-scale 6-DoF task-oriented grasping dataset, 6-DoF Task Grasp (6DTG), which features 4391 cluttered scenes with over 2 million 6-DoF grasp poses. Each grasp is annotated with a specific task, involving 6 tasks and 198 objects in total. Moreover, we propose One-Stage TaskGrasp (OSTG), a strong baseline to address the TO6DGC problem. Our OSTG adopts a task-oriented point selection strategy to detect where to grasp, and a task-oriented grasp generation module to decide how to grasp given a specific task. To evaluate the effectiveness of OSTG, extensive experiments are conducted on 6DTG. The results show that our method outperforms various baselines on multiple metrics. Real robot experiments also verify that our OSTG has a better perception of the task-oriented grasp points and 6-DoF grasp poses. © 2025, CC BY.

关键词： Clutter (information theory)

Dexterous Grasp Transformer

学校读者我要写书评

暂无评论

Dexterous Grasp Transformer

Conference on Computer Vision and Pattern Recognition (CVPR)

作者： Guo-Hao Xu Yi-Lin Wei Dian Zheng Xiao-Ming Wu Wei-Shi Zheng School of Computer Science and Engineering Sun Yat-sen University China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China

ISBN: (数字)9798350353006

ISBN: (纸本)9798350353013

In this work, we propose a novel discriminative frame-work for dexterous grasp generation, named Dexterous Grasp TRansformer (DGTR), capable of predicting a di-verse set of feasible grasp poses by processing the object point cloud with only one forward pass. We formulate dex-terous grasp generation as a set prediction task and design a transformer-based grasping model for it. However, we identify that this set prediction paradigm encounters sev-eral optimization challenges in the field of dexterous grasping and results in restricted performance. To address these issues, we propose progressive strategies for both the training and testing phases. First, the dynamic-static matching training (DSMT) strategy is presented to enhance the opti-mization stability during the training phase. Second, we in-troduce the adversarial-balanced test-time adaptation (AB-TTA) with a pair of adversarial losses to improve grasping quality during the testing phase. Experimental results on the DexGraspNet dataset demonstrate the capability of DGTR to predict dexterous grasp poses with both high quality and diversity. Notably, while keeping high qual-ity, the diversity of grasp poses predicted by DGTR sig-nificantly outperforms previous works in multiple metrics without any data pre-processing. Codes are available at https://***/iSEE-laboratory/DGTR.

关键词： Training Point cloud compression Measurement Grasping Predictive models Transformers Stability analysis

Real-to-Sim Grasp: Rethinking the Gap between Simulation and Real World in Grasp Detection

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Cai, Jia-Feng Chen, Zibo Wu, Xiao-Ming Jiang, Jian-Jian Wei, Yi-Lin Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China

For 6-DoF grasp detection, simulated data is expandable to train more powerful model, but it faces the challenge of the large gap between simulation and real world. Previous works bridge this gap with a sim-to-real way1. However, this way explicitly or implicitly forces the simulated data to adapt to the noisy real data when training grasp detectors, where the positional drift and structural distortion within the camera noise will harm the grasp learning. In this work, we propose a Real-to-Sim framework for 6-DoF Grasp detection, named R2SGrasp, with the key insight of bridging this gap in a real-to-sim way2, which directly bypasses the camera noise in grasp detector training through an inference-time real-to-sim adaption. To achieve this real-to-sim adaptation, our R2SGrasp designs the Real-to-Sim Data Repairer (R2SRepairer) to mitigate the camera noise of real depth maps in data-level, and the Real-to-Sim Feature Enhancer (R2SEnhancer) to enhance real features with precise simulated geometric primitives in feature-level. To endow our framework with the generalization ability, we construct a large-scale simulated dataset cost-efficiently to train our grasp detector, which includes 64,000 RGB-D images with 14.4 million grasp annotations. Sufficient experiments show that R2SGrasp is powerful and our real-to-sim perspective is effective. The real-world experiments further show great generalization ability of R2SGrasp. Project page is available on https://***/R2SGrasp. Copyright © 2024, The Authors. All rights reserved.

关键词：

Dexterous Grasp Transformer

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Xu, Guo-Hao Wei, Yi-Lin Zheng, Dian Wu, Xiao-Ming Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China

In this work, we propose a novel discriminative framework for dexterous grasp generation, named Dexterous Grasp TRansformer (DGTR), capable of predicting a diverse set of feasible grasp poses by processing the object point cloud with only one forward pass. We formulate dexterous grasp generation as a set prediction task and design a transformer-based grasping model for it. However, we identify that this set prediction paradigm encounters several optimization challenges in the field of dexterous grasping and results in restricted performance. To address these issues, we propose progressive strategies for both the training and testing phases. First, the dynamic-static matching training (DSMT) strategy is presented to enhance the optimization stability during the training phase. Second, we introduce the adversarial-balanced test-time adaptation (AB-TTA) with a pair of adversarial losses to improve grasping quality during the testing phase. Experimental results on the DexGraspNet dataset demonstrate the capability of DGTR to predict dexterous grasp poses with both high quality and diversity. Notably, while keeping high quality, the diversity of grasp poses predicted by DGTR significantly outperforms previous works in multiple metrics without any data pre-processing. Codes are available at https://***/iSEE-laboratory/DGTR. © 2024, CC BY.

关键词： Forecasting

EmoDamp: A Model for Speaker Emotion Inference in Dialogues with Bidirectional Emotion Damping Mechanism 6

学校读者我要写书评

暂无评论

EmoDamp: A Model for Speaker Emotion Inference in Dialogues ...

6th International Conference on Algorithms, computing and Artificial intelligence, ACAI 2023

作者： Jiang, Yueqi Sun, Xiao Wang, Jiamin AnHui Province Key Laboratory of Affective Computing and Advanced Intelligent Machine School of Computer Science and Information Engineering Hefei University of Technology Anhui Hefei China Institute of Artificial Intelligence Hefei Comprehensive National Science Center Anhui Hefei China

ISBN: (纸本)9798400709203

Emotion analysis in conversation has been a popular research topic in the natural language processing field. While much of the existing research has focused on emotion recognition in conversation, the emotion inference task in conversation is more challenging due to its specific prerequisites. In this paper, we propose EmoDamp, a novel model for the conversational emotion inference task. By referring to psychological theories, a bidirectional emotion damping network(BED Net) is designed to model the emotion generation mechanism. Furthermore, external commonsense knowledge is also used to supplement the prediction results to make them more consistent with authentic logic. Finally, a conditional random field (CRF) is used to capture the potential constraints of the emotion transfer process in conversation and simulate emotion contagion. Experimental results on two benchmark datasets demonstrate that EmoDamp outperforms the baseline model, achieving improved results on each emotion category and showcasing its effectiveness. © 2023 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： Human computer interaction

VulD-CodeBERT: CodeBERT-Based Vulnerability Detection Model for C/C++ Code

学校读者我要写书评

暂无评论

VulD-CodeBERT: CodeBERT-Based Vulnerability Detection Model ...

International Conference on Communications, Information System and Computer engineering (CISCE)

作者： Zhe Xiong Weiyu Dong School of Computer Science and Artificial Intelligence Zhengzhou University Zhengzhou China State Key Laboratory of Mathematical Engineering and Advanced Computing Information Engineering University Zhengzhou China

ISBN: (数字)9798350353174

ISBN: (纸本)9798350353181

Recently, deep learning techniques have received widespread attention in code vulnerability and vulnerability detection. However, there are some problems with current vulnerability detection techniques, firstly, because of the non-homogeneity of vulnerability categories and quantities, there are omission false positives and overfitting problems for vulnerabilities with a small amount of data. Second, the existing detection methods have limited ability to characterize the code, there is insufficient depth of characterization, does not contain all of its vulnerability characteristics or is interspersed with useless information. Finally, the current research does not use the powerful feature mining capability of pre-trained models and does not employ model fusion to increase the effectiveness of model vulnerability detection. To solve the above problems, this paper proposes a new deep learning based pre-trained model for code vulnerability detection tool VulD-CodeBERT. By fine-tuning CodeBERT, a code adaptation extension for pre-trained models developed by Microsoft, the CodeBERT model can solve the problem of not being able to learn the vulnerability patterns due to insufficient data, thus resulting in missed false alarms. The sequence output from the CodeBERT model are added to a BiLSTM network with an attention mechanism to mine the local key features of the vulnerabilities combined with the global features of the pooler output from the CodeBERT model to deeply mine the dependencies between the vulnerability statements and derive the probability of the existence of the vulnerability. For vulnerability types that are difficult to uncover captured features or the number of samples is not enough, a polynomial plain Bayesian classifier is used for detection, and finally the two models are combined to deeply mine the vulnerability features, thus solving the problem of difficult and small samples, and reducing the leakage rate and false alarm rate. Experimental results sh

关键词： Deep learning Adaptation models Codes Attention mechanisms Computational modeling Feature extraction Polynomials

Expert with Outlier Exposure for Continual Learning of New Diseases

学校读者我要写书评

暂无评论

Expert with Outlier Exposure for Continual Learning of New D...

2022 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2022

作者： Xu, Zhengjing Chen, Kanghao Zheng, Wei-Shi Tan, Zhijun Yang, Xiaobo Wang, Ruixuan Sun Yat-sen University School of Computer Science and Engineering Guangzhou China Key Laboratory of Machine Intelligence and Advanced Computing MOE China State Key Laboratory of Dampness Syndrome of Chinese Medicine Guangzhou China Peng Cheng Laboratory Department of Network Intelligence Shenzhen China

ISBN: (纸本)9781665468190

Current intelligent diagnosis systems struggle to continually learn to diagnose more and more diseases due to catastrophic forgetting of old knowledge when learning new knowledge. Although storing small old data for subsequent continual learning can effectively help alleviate the forgetting issue, the heavy data imbalance between old classes and to-be-learned new classes in classifier training often causes biased prediction towards the new classes just learned by the updated classifier. In this study, an outlier detection technique is novelly applied to train an additional expert classifier for new classes to help alleviate the class imbalance issue and discriminate the learned new classes from old classes during inference (instead of the training phase). Specially, the stored small data of old classes are considered as outliers during training the expert classifier, such that the output probability distributions from the expert classifier are expected to be obviously different between test data of the old classes and those of the new classes. Such difference between old classes and new classes can be used to fine-tune the original output from the updated classifier which is responsible for prediction of all learned (old and new) classes. During inference, a novel ensemble strategy is proposed to combine the predictions from the updated classifier, the expert classifier, and the previously learned old classifier. The proposed learning and inference framework can be easily combined with existing continual learning strategies. Empirical evaluations on three medical image datasets and one natural image dataset show that the proposed framework can effectively improve continual learning performance. © 2022 IEEE.

关键词： Forecasting

SPAFuzz: Web Security Fuzz Testing Tool Based on Fuzz Testing and Genetic Algorithm

学校读者我要写书评

暂无评论

SPAFuzz: Web Security Fuzz Testing Tool Based on Fuzz Testin...

Consumer Electronics and Computer engineering (ICCECE), 2021 IEEE International Conference on

作者： Jiang Lin Weiyu Dong School of Computer Science and Artificial Intelligence Zhengzhou University Zhengzhou China State Key Laboratory of Mathematical Engineering and Advanced Computing Information Engineering University Zhengzhou China

Fuzzing is one of the most effective methods currently available for detecting web vulnerabilities. However, traditional fuzzing suffers from issues such as low coverage and limited diversity in test cases. To address these issues, we propose a way to generate test cases based on dynamic feature libraries and attack models by analysing the characteristics of SQL and XSS. Additionally, we introduce a genetic algorithm to guide the population's evolution by monitoring the state changes of the web application. Furthermore, we improved the selection phase of the genetic algorithm to optimally deform multiple iterations for use cases with low fitness values in order to preserve the population diversity to expand the attack capability of the test cases and based on this the SPAFuzz fuzzy testing tool was implemented. Experimental results show that SPAFuzz is able to detect more network vulnerabilities than Sqlmap and Powerfuzzer.

关键词：

MaintaAvatar: A Maintainable Avatar Based on Neural Radiance Fields by Continual Learning

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Gu, Shengbo Qiu, Yu-Kun Tang, Yu-Ming Wu, Ancong Zheng, Wei-Shi School of Computer Science and Engineering Sun Yat-sen University China Peng Cheng Laboratory Shenzhen China Key Laboratory of Machine Intelligence and Advanced Computing Ministry of Education China

The generation of a virtual digital avatar is a crucial research topic in the field of computer vision. Many existing works utilize Neural Radiance Fields (NeRF) to address this issue and have achieved impressive results. However, previous works assume the images of the training person are available and fixed while the appearances and poses of a subject could constantly change and increase in real-world scenarios. How to update the human avatar but also maintain the ability to render the old appearance of the person is a practical challenge. One trivial solution is to combine the existing virtual avatar models based on NeRF with continual learning methods. However, there are some critical issues in this approach: learning new appearances and poses can cause the model to forget past information, which in turn leads to a degradation in the rendering quality of past appearances, especially color bleeding issues, and incorrect human body poses. In this work, we propose a maintainable avatar (MaintaAvatar) based on neural radiance fields by continual learning, which resolves the issues by utilizing a Global-Local Joint Storage Module and a Pose Distillation Module. Overall, our model requires only limited data collection to quickly fine-tune the model while avoiding catastrophic forgetting, thus achieving a maintainable virtual avatar. The experimental results validate the effectiveness of our MaintaAvatar model. © 2025, CC BY.

关键词： Federated learning