1 code implementation • ECCV 2020 • Kun Ding, Guojin He, Huxiang Gu, Zisha Zhong, Shiming Xiang, Chunhong Pan
State-of-the-art object detectors exploit multi-branch structure and predict objects at several different scales, although substantially boosted accuracy is acquired, low efficiency is inevitable as fragmented structure is hardware unfriendly.
1 code implementation • 31 Mar 2024 • Kun Ding, Haojian Zhang, Qiang Yu, Ying Wang, Shiming Xiang, Chunhong Pan
The idea is realized by exploiting out-of-distribution (OOD) detection to predict whether a sample belongs to a base distribution or a novel distribution and then using the score generated by a dedicated competition based scoring function to fuse the zero-shot and few-shot classifier.
no code implementations • 25 Mar 2024 • Lingdong Shen, Fangxin Shang, Yehui Yang, Xiaoshuang Huang, Shiming Xiang
Extensive experimental validation of SegICL demonstrates a positive correlation between the number of prompt samples and segmentation performance on OOD modalities and tasks.
no code implementations • 24 Mar 2024 • Bolin Ni, Hongbo Zhao, Chenghao Zhang, Ke Hu, Gaofeng Meng, Zhaoxiang Zhang, Shiming Xiang
Existing methods commonly utilize the one-hot labels and randomly initialize the classifier head.
no code implementations • 22 Mar 2024 • Shixiong Xu, Gaofeng Meng, Xing Nie, Bolin Ni, Bin Fan, Shiming Xiang
This intriguing phenomenon, discovered in replay-based Class Incremental Learning (CIL), highlights the imbalanced forgetting of learned classes, as their accuracy is similar before the occurrence of catastrophic forgetting.
no code implementations • 18 Mar 2024 • Kun Ding, Xiaohui Li, Qiang Yu, Ying Wang, Haojian Zhang, Shiming Xiang
Context Optimization (CoOp) has emerged as a simple yet effective technique for adapting CLIP-like vision-language models to downstream image recognition tasks.
1 code implementation • 13 Dec 2023 • Tao Zhang, Kun Ding, Jinyong Wen, Yu Xiong, Zeyu Zhang, Shiming Xiang, Chunhong Pan
Self-supervised learning (SSL) for RGB images has achieved significant success, yet there is still limited research on SSL for infrared images, primarily due to three prominent challenges: 1) the lack of a suitable large-scale infrared pre-training dataset, 2) the distinctiveness of non-iconic infrared images rendering common pre-training tasks like masked image modeling (MIM) less effective, and 3) the scarcity of fine-grained textures making it particularly challenging to learn general image features.
1 code implementation • 11 Dec 2023 • Qi Yang, Xing Nie, Tong Li, Pengfei Gao, Ying Guo, Cheng Zhen, Pengfei Yan, Shiming Xiang
For the first time, our framework explores three types of bilateral entanglements within AVS: pixel entanglement, modality entanglement, and temporal entanglement.
no code implementations • 9 May 2023 • Guangliang Cheng, Yunmeng Huang, Xiangtai Li, Shuchang Lyu, Zhaoyang Xu, Qi Zhao, Shiming Xiang
We first introduce some preliminary knowledge for the change detection task, such as problem definition, datasets, evaluation metrics, and transformer basics, as well as provide a detailed taxonomy of existing algorithms from three different perspectives: algorithm granularity, supervision modes, and learning frameworks in the methodology section.
no code implementations • 17 Jan 2023 • Sen Pei, Jiaxi Sun, Richard Yi Da Xu, Bin Fan, Shiming Xiang, Gaofeng Meng
Generally, existing approaches in dealing with out-of-distribution (OOD) detection mainly focus on the statistical difference between the features of OOD and in-distribution (ID) data extracted by the classifiers.
no code implementations • CVPR 2023 • Xing Nie, Shixiong Xu, Xiyan Liu, Gaofeng Meng, Chunlei Huo, Shiming Xiang
Humans are proficient at continuously acquiring and integrating new knowledge.
1 code implementation • 29 Aug 2022 • Kun Ding, Ying Wang, Pengzhang Liu, Qiang Yu, Haojian Zhang, Shiming Xiang, Chunhong Pan
Inspired by the fact that modeling task relationship by multi-task learning can usually boost performance, we propose a novel method SoftCPT (Soft Context Sharing for Prompt Tuning) to tune pre-trained vision-language models on multiple target few-shot tasks jointly.
2 code implementations • 4 Aug 2022 • Bolin Ni, Houwen Peng, Minghao Chen, Songyang Zhang, Gaofeng Meng, Jianlong Fu, Shiming Xiang, Haibin Ling
Extensive experiments demonstrate that our approach is effective and can be generalized to different video recognition scenarios.
Ranked #8 on Zero-Shot Action Recognition on Kinetics
no code implementations • 28 Jul 2022 • Xing Nie, Bolin Ni, Jianlong Chang, Gaomeng Meng, Chunlei Huo, Zhaoxiang Zhang, Shiming Xiang, Qi Tian, Chunhong Pan
To this end, we propose parameter-efficient Prompt tuning (Pro-tuning) to adapt frozen vision models to various downstream vision tasks.
1 code implementation • 25 Jul 2022 • Sen Pei, Jiaxi Sun, Richard Yi Da Xu, Shiming Xiang, Gaofeng Meng
PoER helps the neural networks to capture label-related features which contain the domain information first in shallow layers and then distills the label-discriminative representations out progressively, enforcing the neural networks to be aware of the characteristic of objects and background which is vital to the generation of domain-invariant features.
no code implementations • CVPR 2022 • Nuo Xu, Jianlong Chang, Xing Nie, Chunlei Huo, Shiming Xiang, Chunhong Pan
Training Deep Neural Networks (DNNs) is inherently subject to sensitive hyper-parameters and untimely feedbacks of performance evaluation.
no code implementations • 21 Dec 2021 • Kailun Wu, Zhangming Chan, Weijie Bian, Lejian Ren, Shiming Xiang, Shuguang Han, Hongbo Deng, Bo Zheng
We further show that such a process is equivalent to adding an adversarial perturbation to the model input, and thereby name our proposed approach as an the Adversarial Gradient Driven Exploration (AGE).
no code implementations • 5 Aug 2021 • Sen Pei, Richard Yi Da Xu, Shiming Xiang, Gaofeng Meng
We compare the proposed method with Unrolled GAN (Metz et al. 2016), BourGAN (Xiao, Zhong, and Zheng 2018), PacGAN (Lin et al. 2018), VEEGAN (Srivastava et al. 2017) and ALI (Dumoulin et al. 2016) on 2D synthetic dataset, and results show that the diversity penalty module can help GAN capture much more modes of the data distribution.
no code implementations • 25 Jun 2021 • Jianbo Liu, Ying Wang, Shiming Xiang, Chunhong Pan
Previous methods for skeleton-based gesture recognition mostly arrange the skeleton sequence into a pseudo picture or spatial-temporal graph and apply deep Convolutional Neural Network (CNN) or Graph Convolutional Network (GCN) for feature extraction.
1 code implementation • 25 May 2021 • Hao He, Xiangtai Li, Yibo Yang, Guangliang Cheng, Yunhai Tong, Lubin Weng, Zhouchen Lin, Shiming Xiang
This module is used to squeeze the object boundary from both inner and outer directions, which contributes to precise mask representation.
1 code implementation • ICCV 2021 • Kun Tian, Chenghao Zhang, Ying Wang, Shiming Xiang, Chunhong Pan
Specifically, KTNet is constructed on a base detector with intrinsic knowledge mining and relational knowledge constraints.
no code implementations • 11 Nov 2020 • Weijie Bian, Kailun Wu, Lejian Ren, Qi Pi, Yujing Zhang, Can Xiao, Xiang-Rong Sheng, Yong-Nan Zhu, Zhangming Chan, Na Mou, Xinchen Luo, Shiming Xiang, Guorui Zhou, Xiaoqiang Zhu, Hongbo Deng
For example, a simple attempt to learn the combination of feature A and feature B <A, B> as the explicit cartesian product representation of new features can outperform previous implicit feature interaction models including factorization machine (FM)-based models and their variations.
no code implementations • AAAI 2020 • Qi Zhang, Jianlong Chang, Gaofeng Meng, Shiming Xiang, Chunhong Pan
To address these issues, we propose a novel framework named Structure Learning Convolution (SLC) that enables to extend the traditional convolutional neural network (CNN) to graph domains and learn the graph structure for traffic forecasting.
Ranked #3 on Traffic Prediction on METR-LA
2 code implementations • CVPR 2020 • Chaoxu Guo, Bin Fan, Qian Zhang, Shiming Xiang, Chunhong Pan
In this paper, we begin by first analyzing the design defects of feature pyramid in FPN, and then introduce a new feature pyramid architecture named AugFPN to address these problems.
1 code implementation • NeurIPS 2019 • Jianlong Chang, Xinbang Zhang, Yiwen Guo, Gaofeng Meng, Shiming Xiang, Chunhong Pan
Neural architecture search (NAS) is inherently subject to the gap of architectures during searching and validating.
1 code implementation • ECCV 2020 • Zhengkai Jiang, Yu Liu, Ceyuan Yang, Jihao Liu, Peng Gao, Qian Zhang, Shiming Xiang, Chunhong Pan
Transferring existing image-based detectors to the video is non-trivial since the quality of frames is always deteriorated by part occlusion, rare pose, and motion blur.
Ranked #23 on Video Object Detection on ImageNet VID
no code implementations • 28 Oct 2019 • Xiyan Liu, Gaofeng Meng, Shiming Xiang, Chunhong Pan
In our model, we decouple character images into style representation and content representation, which facilitates more precise control of these two types of variables, thereby improving the quality of the generated results.
1 code implementation • ICCV 2019 • Yongcheng Liu, Bin Fan, Gaofeng Meng, Jiwen Lu, Shiming Xiang, Chunhong Pan
Point cloud processing is very challenging, as the diverse shapes formed by irregular points are often indistinguishable.
Ranked #22 on 3D Part Segmentation on ShapeNet-Part
no code implementations • 6 May 2019 • Jianlong Chang, Xinbang Zhang, Yiwen Guo, Gaofeng Meng, Shiming Xiang, Chunhong Pan
For network architecture search (NAS), it is crucial but challenging to simultaneously guarantee both effectiveness and efficiency.
no code implementations • 5 May 2019 • Jianlong Chang, Yiwen Guo, Lingfeng Wang, Gaofeng Meng, Shiming Xiang, Chunhong Pan
Traditional clustering methods often perform clustering with low-level indiscriminative representations and ignore relationships between patterns, resulting in slight achievements in the era of deep learning.
15 code implementations • 24 Apr 2019 • Yonghao He, Dezhong Xu, Lifang Wu, Meng Jian, Shiming Xiang, Chunhong Pan
Under the new schema, the proposed method can achieve superior accuracy (WIDER FACE Val/Test -- Easy: 0. 910/0. 896, Medium: 0. 881/0. 865, Hard: 0. 780/0. 770; FDDB -- discontinuous: 0. 973, continuous: 0. 724).
Ranked #6 on Face Detection on FDDB
4 code implementations • CVPR 2019 • Yongcheng Liu, Bin Fan, Shiming Xiang, Chunhong Pan
Specifically, the convolutional weight for local point set is forced to learn a high-level relation expression from predefined geometric priors, between a sampled point from this point set and the others.
Ranked #10 on 3D Point Cloud Classification on ModelNet40-C
no code implementations • ICCV 2019 • Chaoxu Guo, Bin Fan, Jie Gu, Qian Zhang, Shiming Xiang, Veronique Prinet, Chunhong Pan
Instead of relying on optical flow, this paper proposes a novel module called Progressive Sparse Local Attention (PSLA), which establishes the spatial correspondence between features across frames in a local region with progressively sparser stride and uses the correspondence to propagate features.
1 code implementation • NeurIPS 2018 • Jianlong Chang, Jie Gu, Lingfeng Wang, Gaofeng Meng, Shiming Xiang, Chunhong Pan
Convolutional neural networks (CNNs) are inherently subject to invariable filters that can only aggregate local inputs with the same topological structures.
no code implementations • 23 Nov 2018 • Yukang Chen, Gaofeng Meng, Qian Zhang, Xinbang Zhang, Liangchen Song, Shiming Xiang, Chunhong Pan
Here our goal is to automatically find a compact neural network model with high performance that is suitable for mobile devices.
1 code implementation • 16 Sep 2018 • Yongcheng Liu, Lu Sheng, Jing Shao, Junjie Yan, Shiming Xiang, Chunhong Pan
Specifically, given the image-level annotations, (1) we first develop a weakly-supervised detection (WSD) model, and then (2) construct an end-to-end multi-label image classification framework augmented by a knowledge distillation module that guides the classification model by the WSD model according to the class-level predictions for the whole image and the object-level visual features for object RoIs.
Ranked #9 on Multi-Label Classification on NUS-WIDE
no code implementations • ECCV 2018 • Gaofeng MENG, Yuanqi SU, Ying Wu, Shiming Xiang, Chunhong Pan
This paper proposes a segment-free method for geometric rectification of a distorted document image captured by a hand-held camera.
1 code implementation • 1 Aug 2018 • Yukang Chen, Gaofeng Meng, Qian Zhang, Shiming Xiang, Chang Huang, Lisen Mu, Xinggang Wang
To address this issue, we propose the Reinforced Evolutionary Neural Architecture Search (RE- NAS), which is an evolutionary method with the reinforced mutation for NAS.
1 code implementation • 30 Jul 2018 • Yongcheng Liu, Bin Fan, Lingfeng Wang, Jun Bai, Shiming Xiang, Chunhong Pan
Specifically, for confusing manmade objects, ScasNet improves the labeling coherence with sequential global-to-local contexts aggregation.
no code implementations • 14 Dec 2017 • Bin Fan, Qingqun Kong, Xinchao Wang, Zhiheng Wang, Shiming Xiang, Chunhong Pan, Pascal Fua
To obtain a comprehensive evaluation, we choose to include both float type features and binary ones.
1 code implementation • ICCV 2017 • Jianlong Chang, Lingfeng Wang, Gaofeng Meng, Shiming Xiang, Chunhong Pan
The main challenge is that the ground-truth similarities are unknown in image clustering.
Ranked #8 on Image Clustering on Tiny-ImageNet
no code implementations • CVPR 2017 • Cheng Da, Shibiao Xu, Kun Ding, Gaofeng Meng, Shiming Xiang, Chunhong Pan
(2) A multi-integer-embedding is employed for compressing the whole database, which is modeled by binary sparse representation with fixed sparsity.
no code implementations • 14 Feb 2016 • Bin Fan, Qingqun Kong, Wei Sui, Zhiheng Wang, Xinchao Wang, Shiming Xiang, Chunhong Pan, Pascal Fua
Binary features have been incrementally popular in the past few years due to their low memory footprints and the efficient computation of Hamming distance between binary descriptors.
no code implementations • ICCV 2015 • Gaofeng Meng, Zuming Huang, Yonghong Song, Shiming Xiang, Chunhong Pan
In this paper, we propose an efficient method for accurate extraction of these virtual visual cues from a curved document image.
no code implementations • ICCV 2015 • Feihu Zhang, Longquan Dai, Shiming Xiang, Xiaopeng Zhang
In our SGF, we use the tree distance on the segment graph to define the internal weight function of the filtering kernel, which enables the filter to smooth out high-contrast details and textures while preserving major image structures very well.
no code implementations • 25 Aug 2015 • Guangliang Cheng, Feiyun Zhu, Shiming Xiang, Chunhong Pan
Finally, to overcome the ineffectiveness of current methods in the road intersection, a fitting based road centerline connection algorithm is proposed.
no code implementations • 18 Nov 2014 • Cuicui Kang, Shengcai Liao, Yonghao He, Jian Wang, Wenjia Niu, Shiming Xiang, Chunhong Pan
A new approach to the problem has been raised which intends to match features of different modalities directly.
no code implementations • 12 Sep 2014 • Feiyun Zhu, Bin Fan, Xinliang Zhu, Ying Wang, Shiming Xiang, Chunhong Pan
Subset selection from massive data with noised information is increasingly popular for various applications.
no code implementations • CVPR 2014 • Gaofeng Meng, Ying Wang, Shenquan Qu, Shiming Xiang, Chunhong Pan
Document images captured by a digital camera often suffer from serious geometric distortions.
no code implementations • 19 Mar 2014 • Feiyun Zhu, Ying Wang, Shiming Xiang, Bin Fan, Chunhong Pan
With this constraint, our method can learn a compact space, where highly similar pixels are grouped to share correlated sparse representations.
no code implementations • 13 Mar 2014 • Feiyun Zhu, Ying Wang, Bin Fan, Gaofeng Meng, Shiming Xiang, Chunhong Pan
Hyperspectral unmixing, the process of estimating a common set of spectral bases and their corresponding composite percentages at each pixel, is an important task for hyperspectral analysis, visualization and understanding.
no code implementations • 31 May 2013 • Ying Wang, Chunhong Pan, Shiming Xiang, Feiyun Zhu
In addition, with sparsity constraints, our model can naturally generate sparse abundances.