1 code implementation • 18 May 2024 • Xingyu Miao, Haoran Duan, Varun Ojha, Jun Song, Tejal Shah, Yang Long, Rajiv Ranjan
In this work, we propose a novel Trajectory Score Matching (TSM) method that aims to solve the pseudo ground truth inconsistency problem caused by the accumulated error in Interval Score Matching (ISM) when using the Denoising Diffusion Implicit Models (DDIM) inversion process.
1 code implementation • 17 May 2024 • Rui Sun, Yumin Zhang, Tejal Shah, Jiahao Sun, Shuoying Zhang, Wenqi Li, Haoran Duan, Bo Wei, Rajiv Ranjan
With impressive achievements made, artificial intelligence is on the path forward to artificial general intelligence.
no code implementations • 14 Mar 2024 • Fan Wan, Xingyu Miao, Haoran Duan, Jingjing Deng, Rui Gao, Yang Long
With increasing concerns over data privacy and model copyrights, especially in the context of collaborations between AI service providers and data owners, an innovative SG-ZSL paradigm is proposed in this work.
1 code implementation • 13 Feb 2024 • Chenghao Xiao, Zhuoxu Huang, Danlu Chen, G Thomas Hudson, Yizhi Li, Haoran Duan, Chenghua Lin, Jie Fu, Jungong Han, Noura Al Moubayed
To our knowledge, this is the first representation learning method devoid of traditional language models for understanding sentence and document semantics, marking a stride closer to human-like textual comprehension.
1 code implementation • 2 Feb 2024 • Xingyu Miao, Yang Bai, Haoran Duan, Fan Wan, Yawen Huang, Yang Long, Yefeng Zheng
Most of the existing works on arbitrary 3D NeRF style transfer required retraining on each single style condition.
no code implementations • 10 Jan 2024 • Xingyu Miao, Yang Bai, Haoran Duan, Yawen Huang, Fan Wan, Yang Long, Yefeng Zheng
The goal of our work is to generate high-quality novel views from monocular videos of complex and dynamic scenes.
1 code implementation • 25 Sep 2023 • Lei Xiang, Yuan Zhou, Haoran Duan, Yang Long
To address these issues, we propose a novel Dual Feature Augmentation Network (DFAN), which comprises two feature augmentation modules, one for visual features and the other for semantic features.
1 code implementation • 23 Sep 2023 • Hantao Zhou, Rui Yang, Yachao Zhang, Haoran Duan, Yawen Huang, Runze Hu, Xiu Li, Yefeng Zheng
More precisely, our approach (1) introduces deformation perception, enabling the model to adaptively sample object features; (2) proposes a Dual-axial Aggregation Transformer (DAT) to adeptly model long-range dependencies, thereby achieving global perception; and (3) devises a Cross-task Interaction Transformer (CIT) that facilitates interaction between the classification and localization branches, thus aligning the two tasks.
1 code implementation • 14 Aug 2023 • Xingyu Miao, Yang Bai, Haoran Duan, Yawen Huang, Fan Wan, Xinxing Xu, Yang Long, Yefeng Zheng
Nevertheless, the dynamic cost volume inevitably generates extra occlusions and noise, thus we alleviate this by designing a fusion module that makes static and dynamic cost volumes compensate for each other.
no code implementations • 2 Nov 2022 • Peng Zhang, Yawen Huang, Bingzhang Hu, Shizheng Wang, Haoran Duan, Noura Al Moubayed, Yefeng Zheng, Yang Long
Reinforcement Learning (RL)-based control system has received considerable attention in recent decades.
no code implementations • 23 Feb 2022 • Rui Gao, Fan Wan, Daniel Organisciak, Jiyao Pu, Junyan Wang, Haoran Duan, Peng Zhang, Xingsong Hou, Yang Long
Considering the increasing concerns about data copyright and privacy issues, we present a novel Absolute Zero-Shot Learning (AZSL) paradigm, i. e., training a classifier with zero real data.
no code implementations • 31 Aug 2021 • Haoran Duan, Fan Wan, Rui Sun, Zeyu Wang, Varun Ojha, Yu Guan, Hubert P. H. Shum, Bingzhang Hu, Yang Long
Our method achieved competitive performance in semi-supervised learning approaches on these crowd counting datasets.
1 code implementation • 25 Mar 2021 • Rui Wang, Zhihua Wei, Haoran Duan, Shouling Ji, Yang Long, Zhen Hong
Compared with hand-designed approaches, neural architecture search (NAS) appears as a practical technique in automating the manual architecture design process and has attracted increasing interest in spoken language processing tasks such as speaker recognition.
no code implementations • 9 Aug 2020 • Haoran Duan, Shidong Wang, Yu Guan
To obtain the appropriate crowd representation, in this work we proposed SOFA-Net(Second-Order and First-order Attention Network): second-order statistics were extracted to retain selectivity of the channel-wise spatial information for dense heads while first-order statistics, which can enhance the feature discrimination for the heads' areas, were used as complementary information.