1 code implementation • 22 Apr 2024 • Junyu Gao, Da Zhang, Xuelong Li
Then, based on the theory, we design a DPD algorithm which is composed by a training paradigm and proxy domain generator to enhance the domain generalization of the confidence-threshold learner.
1 code implementation • 19 Jan 2024 • Junyu Gao, Liangliang Zhao, Xuelong Li
Considering the absence of a dataset for this task, a large-scale Dataset (NWPU-MOC) is collected, consisting of 3, 416 scenes with a resolution of 1024 $\times$ 1024 pixels, and well-annotated using 14 fine-grained object categories.
1 code implementation • 12 Jan 2024 • Haoxuan Ding, Junyu Gao, Yuan Yuan, Qi Wang
Meanwhile, the proposed SamLP has great few-shot and zero-shot learning ability, which shows the potential of transferring vision foundation model.
no code implementations • 22 Nov 2023 • Junyu Gao, Xuan Yao, Changsheng Xu
Then, these components are adaptively accumulated to pinpoint a concordant direction for fast model adaptation.
1 code implementation • 6 Nov 2023 • Shengkai Sun, Daizong Liu, Jianfeng Dong, Xiaoye Qu, Junyu Gao, Xun Yang, Xun Wang, Meng Wang
In this manner, our framework is able to learn the unified representations of uni-modal or multi-modal skeleton input, which is flexible to different kinds of modality input for robust action understanding in practical cases.
no code implementations • 12 Oct 2023 • Junyu Gao, Xinhong Ma, Changsheng Xu
Despite the great progress of unsupervised domain adaptation (UDA) with the deep neural networks, current UDA models are opaque and cannot provide promising explanations, limiting their applications in the scenarios that require safe and controllable model decisions.
no code implementations • 11 Jul 2023 • Sikai Bai, Shuaicheng Li, Weiming Zhuang, Jie Zhang, Song Guo, Kunlin Yang, Jun Hou, Shuai Zhang, Junyu Gao, Shuai Yi
Theoretically, we show the convergence guarantee of the dual regulators.
no code implementations • 5 Jul 2023 • Jie Fu, Junyu Gao, Changsheng Xu
In this paper, to balance the feature learning processes of different modalities, a dynamic gradient modulation (DGM) mechanism is explored, where a novel and effective metric function is designed to measure the imbalanced feature learning between audio and visual modalities.
no code implementations • 17 May 2023 • Hao Yang, Junyu Gao, Yuan Yuan, Xuelong Li
Anomaly detection in temporal data from sensors under aviation scenarios is a practical but challenging task: 1) long temporal data is difficult to extract contextual information with temporal correlation; 2) the anomalous data are rare in time series, causing normal/abnormal imbalance in anomaly detection, making the detector classification degenerate or even fail.
1 code implementation • CVPR 2023 • Junyu Gao, Mengyuan Chen, Changsheng Xu
We argue that, for an event residing in one modality, the modality itself should provide ample presence evidence of this event, while the other complementary modality is encouraged to afford the absence evidence as a reference signal.
no code implementations • CVPR 2023 • Mengyuan Chen, Junyu Gao, Changsheng Xu
Targeting at recognizing and localizing action instances with only video-level labels during training, Weakly-supervised Temporal Action Localization (WTAL) has achieved significant progress in recent years.
Open Set Learning Weakly-supervised Temporal Action Localization +1
1 code implementation • Conference 2022 • Wei Lin, Kunlin Yang, Xinzhu Ma, Junyu Gao, Lingbo Liu, Shinan Liu, Jun Hou, Shuai Yi, Antoni B. Chan
Here we propose a scale-sensitive generalized loss to tackle this problem.
Ranked #7 on Object Counting on FSC147
no code implementations • 2 Dec 2022 • Qi Wang, Juncheng Wang, Junyu Gao, Yuan Yuan, Xuelong Li
The mainstream crowd counting methods regress density map and integrate it to obtain counting results.
no code implementations • 14 Aug 2022 • PengYu Chen, Junyu Gao, Yuan Yuan, Qi Wang
RGB-Thermal (RGB-T) crowd counting is a challenging task, which uses thermal images as complementary information to RGB images to deal with the decreased performance of unimodal RGB-based methods in scenes with low-illumination or similar backgrounds.
no code implementations • 12 Jun 2022 • Juncheng Wang, Junyu Gao, Yuan Yuan, Qi Wang
The core reason of intrinsic scale shift being one of the most essential issues in crowd localization is that it is ubiquitous in crowd scenes and makes scale distribution chaotic.
no code implementations • 22 May 2022 • Yufan Hu, Junyu Gao, Changsheng Xu
Most existing state-of-the-art video classification methods assume that the training data obey a uniform distribution.
1 code implementation • 4 Apr 2022 • Ziyue Wu, Junyu Gao, Shucheng Huang, Changsheng Xu
Then, a commonsense-aware interaction module is designed to obtain bridged visual and text features by utilizing the learned commonsense concepts.
1 code implementation • CVPR 2022 • Junyu Gao, Mengyuan Chen, Changsheng Xu
We target at the task of weakly-supervised action localization (WSAL), where only video-level action labels are available during model training.
2 code implementations • CVPR 2022 • Tao Han, Lei Bai, Junyu Gao, Qi Wang, Wanli Ouyang
Instead of relying on the Multiple Object Tracking (MOT) techniques, we propose to solve the problem by decomposing all pedestrians into the initial pedestrians who existed in the first frame and the new pedestrians with separate identities in each following frame.
no code implementations • 1 Dec 2021 • Wei Wang, Junyu Gao, Changsheng Xu
With this in mind, we design a unified causal framework to learn the deconfounded object-relevant association for more accurate and robust video object grounding.
no code implementations • 28 Oct 2021 • Junyu Gao, Maoguo Gong, Xuelong Li
The second is an audio CNN for encoding Log Mel-Spectrogram of audio signals.
no code implementations • 10 Oct 2021 • Qi Wang, Tao Han, Junyu Gao, Yuan Yuan, Xuelong Li
The rapid development in visual crowd analysis shows a trend to count people by positioning or even detecting, rather than simply summing a density map.
no code implementations • 12 Sep 2021 • Qi Wang, Sikai Bai, Junyu Gao, Yuan Yuan, Xuelong Li
In addition, due to domain gaps between different datasets, the performance is dramatically decreased when re-ID models pre-trained on label-rich datasets (source domain) are directly applied to other unlabeled datasets (target domain).
1 code implementation • 2 Aug 2021 • Junyu Gao, Maoguo Gong, Xuelong Li
To this end, we propose a Dilated Convolutional Swin Transformer (DCST) for congested crowd scenes.
1 code implementation • 19 Jul 2021 • Dawei Du, Longyin Wen, Pengfei Zhu, Heng Fan, QinGhua Hu, Haibin Ling, Mubarak Shah, Junwen Pan, Ali Al-Ali, Amr Mohamed, Bakour Imene, Bin Dong, Binyu Zhang, Bouchali Hadia Nesma, Chenfeng Xu, Chenzhen Duan, Ciro Castiello, Corrado Mencar, Dingkang Liang, Florian Krüger, Gennaro Vessio, Giovanna Castellano, Jieru Wang, Junyu Gao, Khalid Abualsaud, Laihui Ding, Lei Zhao, Marco Cianciotta, Muhammad Saqib, Noor Almaadeed, Omar Elharrouss, Pei Lyu, Qi Wang, Shidong Liu, Shuang Qiu, Siyang Pan, Somaya Al-Maadeed, Sultan Daud Khan, Tamer Khattab, Tao Han, Thomas Golda, Wei Xu, Xiang Bai, Xiaoqing Xu, Xuelong Li, Yanyun Zhao, Ye Tian, Yingnan Lin, Yongchao Xu, Yuehan Yao, Zhenyu Xu, Zhijian Zhao, Zhipeng Luo, Zhiwei Wei, Zhiyuan Zhao
Crowd counting on the drone platform is an interesting topic in computer vision, which brings new challenges such as small object inference, background clutter and wide viewpoint.
1 code implementation • 19 Jul 2021 • Haopeng Li, Lingbo Liu, Kunlin Yang, Shinan Liu, Junyu Gao, Bin Zhao, Rui Zhang, Jun Hou
Video crowd localization is a crucial yet challenging task, which aims to estimate exact locations of human heads in the given crowded videos.
no code implementations • 23 Mar 2021 • Xuan Ma, Xiaoshan Yang, Junyu Gao, Changsheng Xu
However, these data streams are multi-source and heterogeneous, containing complex temporal structures with local contextual and global temporal aspects, which makes the feature learning and data joint utilization challenging.
no code implementations • ICCV 2021 • Xinhong Ma, Junyu Gao, Changsheng Xu
This paper proposes a new paradigm for unsupervised domain adaptation, termed as Active Universal Domain Adaptation (AUDA), which removes all label set assumptions and aims for not only recognizing target samples from source classes but also inferring those from target-private classes by using active learning to annotate a small budget of target data.
no code implementations • ICCV 2021 • Junyu Gao, Changsheng Xu
To tackle this issue, we replace the cross-modal interaction module with a cross-modal common space, in which moment-query alignment is learned and efficient moment search can be performed.
1 code implementation • 8 Dec 2020 • Junyu Gao, Tao Han, Qi Wang, Yuan Yuan, Xuelong Li
Furthermore, to improve the segmentation quality for different density regions, we present a differentiable Binarization Module (BM) to output structured instance maps.
1 code implementation • NeurIPS 2020 • Tao Han, Junyu Gao, Yuan Yuan, Qi Wang
In this paper, we combine both to propose an Unsupervised Semantic Aggregation and Deformable Template Matching (USADTM) framework for SSL, which strives to improve the classification performance with few labeled data and then reduce the cost in data annotating.
1 code implementation • Proceedings of the AAAI Conference on Artificial Intelligence 2019 • Junyu Gao, Tianzhu Zhang, Changsheng Xu
To effectively leverage the knowledge graph, we design a novel Two-Stream Graph Convolutional Network (TS-GCN) consisting of a classifier branch and an instance branch.
Ranked #5 on Zero-Shot Action Recognition on Olympics