no code implementations • 22 Apr 2024 • Kun Sun, Rong Wang
Data-driven approaches have revolutionized scientific research.
no code implementations • 18 Apr 2024 • Rong Wang, Kun Sun
This study employs deep learning techniques to explore four speaker profiling tasks on the TIMIT dataset, namely gender classification, accent classification, age estimation, and speaker identification, highlighting the potential and challenges of multi-task learning versus single-task models.
no code implementations • 27 Mar 2024 • Kun Sun
Our approach underscores the potential of these metrics to advance our comprehension of how humans understand and process language, ultimately leading to a better understanding of language comprehension and processing.
no code implementations • 23 Mar 2024 • Kun Sun, Rong Wang
Our results indicate that these computational sentence-level metrics are exceptionally effective at predicting and elucidating the processing difficulties encountered by readers in comprehending sentences as a whole across a variety of languages.
no code implementations • 22 Mar 2024 • Kun Sun, Rong Wang, Haitao Liu, Anders Søgaard
Evaluations have revealed that factors such as scaling, training types, architectures and other factors profoundly impact the performance of LLMs.
1 code implementation • 14 Jun 2023 • Xiao He, Chang Tang, Xinwang Liu, Wei zhang, Kun Sun, Jiangfeng Xu
S2ADet comprises a hyperspectral information decoupling (HID) module, a two-stream feature extraction network, and a one-stage detection head.
1 code implementation • 31 May 2023 • Zhenglai Li, Chang Tang, Xianju Li, Weiying Xie, Kun Sun, Xinzhong Zhu
Specifically, an online uncertainty estimation branch is constructed to model the pixel-wise uncertainty, which is supervised by the difference between predicted change maps and corresponding ground truth during the training process.
1 code implementation • ICCV 2023 • Qiao Wu, Jiaqi Yang, Kun Sun, Chu'ai Zhang, Yanning Zhang, Mathieu Salzmann
Specifically, we introduce two cycle-consistency strategies for supervision: 1) Self tracking cycles, which leverage labels to help the model converge better in the early stages of training; 2) forward-backward cycles, which strengthen the tracker's robustness to motion variations and the template noise caused by the template update strategy.
1 code implementation • 28 Mar 2022 • Zhi Chen, Kun Sun, Fan Yang, Wenbing Tao
In this paper, we present a second order spatial compatibility (SC^2) measure based method for efficient and robust point cloud registration (PCR), called SC^2-PCR.
no code implementations • 8 Jan 2022 • Qi Qi, Kunqian Li, Haiyong Zheng, Xiang Gao, Guojia Hou, Kun Sun
In this paper, we propose a novel underwater image enhancement network, called SGUIE-Net, in which we introduce semantic information as high-level guidance across different images that share common semantic regions.
no code implementations • CVPR 2022 • Zhi Chen, Kun Sun, Fan Yang, Wenbing Tao
In this paper, we present a second order spatial compatibility (SC^2) measure based method for efficient and robust point cloud registration (PCR), called SC^2-PCR.
Ranked #1 on Point Cloud Registration on FP-O-H
no code implementations • 21 Aug 2021 • Jiaming Mu, Binghui Wang, Qi Li, Kun Sun, Mingwei Xu, Zhuotao Liu
We also evaluate the effectiveness of our attack under two defenses: one is well-designed adversarial graph detector and the other is that the target GNN model itself is equipped with a defense to prevent adversarial graph generation.
no code implementations • 10 Feb 2021 • Fengting Li, Xuankai Liu, Xiaoli Zhang, Qi Li, Kun Sun, Kang Li
Particularly, the localized adversarial examples only perturb a small and contiguous region of the target object, so that they are robust and effective in both digital and physical worlds.
no code implementations • 23 May 2018 • Tao Xu, Kun Sun, Wenbing Tao
In this paper, we proposed a GPU accelerated image matching method with improved Cascade Hashing.
no code implementations • 21 Dec 2016 • Kun Sun, Wenbing Tao
Accuracy and efficiency are two key problems in large scale incremental Structure from Motion (SfM).
no code implementations • 20 Jul 2016 • Yun Gu, Guang-Zhong Yang, Jie Yang, Kun Sun
The proposed method is comprised of three stages, the frame smoothing, spatial-temporal embedding and final classification.
no code implementations • CVPR 2014 • Wenbing Tao, Kun Sun
The probabilistic methods based on Symmetrical Gauss Mixture Model (SGMM) have achieved great success in point sets registration, but are seldom used to find the correspondences between two images due to the complexity of the non-rigid transformation and too many outliers.