1 code implementation • ECCV 2020 • Aoxiang Fan, Xingyu Jiang, Yang Wang, Junjun Jiang, Jiayi Ma
Geometric estimation from image point correspondences is the core procedure of many 3D vision problems, which is prevalently accomplished by random sampling techniques.
no code implementations • 31 Mar 2025 • Zizhuo Li, Yifan Lu, Linfeng Tang, Shihua Zhang, Jiayi Ma
This prospective study proposes CoMatch, a novel semi-dense image matcher with dynamic covisibility awareness and bilateral subpixel accuracy.
no code implementations • 30 Mar 2025 • Linfeng Tang, Chunyu Li, Guoqing Wang, Yixuan Yuan, Jiayi Ma
This work presents a \textbf{D}egradation and \textbf{S}emantic \textbf{P}rior dual-guided framework for degraded image \textbf{Fusion} (\textbf{DSPFusion}), utilizing degradation priors and high-quality scene semantic priors restored via diffusion models to guide both information recovery and fusion in a unified model.
no code implementations • 30 Mar 2025 • Linfeng Tang, Yeda Wang, Zhanchuan Cai, Junjun Jiang, Jiayi Ma
Current image fusion methods struggle to address the composite degradations encountered in real-world imaging scenarios and lack the flexibility to accommodate user-specific requirements.
no code implementations • 30 Mar 2025 • Linfeng Tang, Yeda Wang, Meiqi Gong, Zizhuo Li, Yuxin Deng, Xunpeng Yi, Chunyu Li, Han Xu, Hao Zhang, Jiayi Ma
Compared to images, videos better align with real-world acquisition scenarios and possess valuable temporal cues.
no code implementations • 23 Mar 2025 • Xiang Fang, Shihua Zhang, Hao Zhang, Tao Lu, Huabing Zhou, Jiayi Ma
Two-view correspondence learning aims to discern true and false correspondences between image pairs by recognizing their underlying different information.
no code implementations • 14 Mar 2025 • Yansheng Li, Yuning Wu, Gong Cheng, Chao Tao, Bo Dang, Yu Wang, Jiahao Zhang, Chuge Zhang, Yiting Liu, Xu Tang, Jiayi Ma, Yongjun Zhang
To address this limitation, we introduce the Million-scale finE-grained geospatial scEne classification dataseT (MEET), which contains over 1. 03 million zoom-free remote sensing scene samples, manually annotated into 80 fine-grained categories.
no code implementations • 24 Feb 2025 • Wendi Liu, Pei Yang, Wenhui Hong, Xiaoguang Mei, Jiayi Ma
We select two of the most widely used dense prediction tasks: semantic segmentation and change detection, and generate datasets suitable for these tasks.
1 code implementation • 31 Oct 2024 • Hao Zhang, Lei Cao, Jiayi Ma
Second, by embedding the combination of the text and zero-shot location model into the diffusion fusion process, a text-controlled fusion re-modulation strategy is developed.
no code implementations • 16 Oct 2024 • Pengwei Liang, Junjun Jiang, Qing Ma, Xianming Liu, Jiayi Ma
To address these limitations, we introduce DeFusion++, a novel framework that leverages self-supervised learning (SSL) to enhance the versatility of feature representation for different image fusion tasks.
1 code implementation • 9 May 2024 • Yuan Gao, Weizhong Zhang, Wenhan Luo, Lin Ma, Jin-Gang Yu, Gui-Song Xia, Jiayi Ma
We aim at exploiting additional auxiliary labels from an independent (auxiliary) task to boost the primary task performance which we focus on, while preserving a single task inference cost of the primary task.
no code implementations • 17 Apr 2024 • Jiayang Li, Junjun Jiang, Pengwei Liang, Jiayi Ma
Instead of being driven by downstream tasks, our model utilizes a pretrained encoder from Masked Autoencoders (MAE), which facilities the omni features extraction for low-level reconstruction and high-level vision tasks, to obtain perception friendly features with a low cost.
1 code implementation • CVPR 2024 • Xunpeng Yi, Han Xu, Hao Zhang, Linfeng Tang, Jiayi Ma
Through the text semantic encoder and semantic interaction fusion decoder, Text-IF is accessible to the all-in-one infrared and visible image degradation-aware processing and the interactive flexible fusion outcomes.
1 code implementation • 20 Mar 2024 • Linshan Wu, Zhun Zhong, Jiayi Ma, Yunchao Wei, Hao Chen, Leyuan Fang, Shutao Li
Based on the label distributions, we leverage the GMM to generate high-quality pseudo labels for more reliable supervision.
Weakly supervised Semantic Segmentation
Weakly-Supervised Semantic Segmentation
1 code implementation • 11 Mar 2024 • Guobao Xiao, Jun Yu, Jiayi Ma, Deng-Ping Fan, Ling Shao
The principle of LSC is to preserve the latent semantic consensus in both data points and model hypotheses.
no code implementations • 16 Jan 2024 • Haibin Zhou, Huabing Zhou, Jun Chang, Tao Lu, Jiayi Ma
Therefore, accurate lane modeling is essential to align prediction results closely with the environment.
1 code implementation • CVPR 2024 • Hao Zhang, Linfeng Tang, Xinyu Xiang, Xuhui Zuo, Jiayi Ma
The GRM utilizes the light-invariant high-contrast characteristics of the infrared modality as the central target distribution and constructs a multi-level conditional adversarial sample set to enable continuous controlled brightness enhancement of visible images.
no code implementations • CVPR 2024 • Yang Yu, Erting Pan, Xinya Wang, Yuheng Wu, Xiaoguang Mei, Jiayi Ma
By integrating unmixing this work maps unpaired HSI and RGB data to a low-dimensional abundance space greatly alleviating the difficulty of generating high-dimensional samples.
1 code implementation • CVPR 2024 • Shihua Zhang, Zizhuo Li, Yuan Gao, Jiayi Ma
Specifically we first decompose the rough motion field that is contaminated by false matches into several different sub-fields which are highly smooth and contain the main energy of the original field.
1 code implementation • CVPR 2024 • Hao Zhang, Xuhui Zuo, Jie Jiang, Chunchao Guo, Jiayi Ma
Then the cascade of IGM-Att and PC-Att couples image fusion and semantic segmentation tasks implicitly bringing vision-related and semantics-related features into closer alignment.
no code implementations • 25 Dec 2023 • Hao Wang, Huabing Zhou, Yanduo Zhang, Tao Lu, Jiayi Ma
Scene text spotting is essential in various computer vision applications, enabling extracting and interpreting textual information from images.
no code implementations • ICCV 2023 • Xunpeng Yi, Han Xu, Hao Zhang, Linfeng Tang, Jiayi Ma
Therefore, Diff-Retinex formulates the low-light image enhancement problem into Retinex decomposition and conditional image generation.
1 code implementation • 11 Jul 2023 • Yuxin Deng, Jiayi Ma
In order to facilitate the learning of matching and filtering, we inject the similarity of descriptors and relative positions into cross- and self-attention score, respectively.
no code implementations • 4 Jul 2023 • Zizhuo Li, Jiayi Ma
Then, our Matchable Keypoint-Assisted Context Aggregation Module regards sampled informative keypoints as message bottlenecks and thus constrains each keypoint only to retrieve favorable contextual information from intra- and inter- matchable keypoints, evading the interference of irrelevant and redundant connectivity with non-repeatable ones.
1 code implementation • 24 Mar 2023 • Pengwei Liang, Junjun Jiang, Xianming Liu, Jiayi Ma
We demonstrate the effectiveness of transformer properties in improving the perceptual quality while not sacrificing the quantitative scores (PSNR) over the most competitive models, such as Uformer, Restormer, and NAFNet, on defocus deblurring and motion deblurring tasks.
no code implementations • 19 Jan 2023 • Jun Yue, Leyuan Fang, Shaobo Xia, Yue Deng, Jiayi Ma
In specific, instead of converting multi-channel images into single-channel data in existing fusion methods, we create the multi-channel data distribution with a denoising network in a latent space with forward and reverse diffusion process.
1 code implementation • CVPR 2023 • Jinsheng Xiao, Yuanxu Wu, Yunhua Chen, Shurui Wang, Zhongyuan Wang, Jiayi Ma
We find that context information from the long-term frame and temporal information from the short-term frame are two useful cues for video small object detection.
1 code implementation • CVPR 2023 • Linshan Wu, Zhun Zhong, Leyuan Fang, Xingxin He, Qiang Liu, Jiayi Ma, Hao Chen
Our AGMM can effectively endow reliable supervision for unlabeled pixels based on the distributions of labeled and unlabeled pixels.
1 code implementation • CVPR 2023 • Yifan Lu, Jiayi Ma, Leyuan Fang, Xin Tian, Junjun Jiang
This enables the application of Gaussian processes to a wide range of real data, which are often large-scale and contaminated by outliers.
no code implementations • 22 Nov 2022 • Bo Dang, Yansheng Li, Yongjun Zhang, Jiayi Ma
Semi-supervised semantic segmentation focuses on the exploration of a small amount of labeled data and a large amount of unlabeled data, which is more in line with the demands of real-world image understanding applications.
no code implementations • 16 May 2022 • Yuxin Deng, Jiayi Ma
Deep-learning-based local feature extraction algorithms that combine detection and description have made significant progress in visible image matching.
no code implementations • 14 Mar 2022 • Youming Deng, Yansheng Li, Yongjun Zhang, Xiang Xiang, Jian Wang, Jingdong Chen, Jiayi Ma
After the autonomous partition of coarse and fine predicates, the model is first trained on the coarse predicates and then learns the fine predicates.
no code implementations • CVPR 2022 • Han Xu, Jiayi Ma, Jiteng Yuan, Zhuliang Le, Wei Liu
Specifically, for image registration, we solve the bottlenecks of defining registration metrics applicable for multi-modal images and facilitating the network convergence.
no code implementations • CVPR 2022 • Aoxiang Fan, Jiayi Ma, Xin Tian, Xiaoguang Mei, Wei Liu
In this paper, we explore a new type of extrinsic method to directly align two geometric shapes with point-to-point correspondences in ambient space by recovering a deformation, which allows more continuous and smooth maps to be obtained.
1 code implementation • CVPR 2022 • Luanyuan Dai, Yizhang Liu, Jiayi Ma, Lifang Wei, Taotao Lai, Changcai Yang, Riqing Chen
However, most such works ignore similar sparse semantics information between two given images and cannot capture local topology among correspondences well.
no code implementations • 27 Nov 2021 • Qing Ma, Junjun Jiang, Xianming Liu, Jiayi Ma
Learning-based methods usually use a convolutional neural network (CNN) to learn the implicit priors of HSIs.
no code implementations • 16 Sep 2021 • Yuanzhi Wang, Tao Lu, Yanduo Zhang, Junjun Jiang, JiaMing Wang, Zhongyuan Wang, Jiayi Ma
Recently, face super-resolution (FSR) methods either feed whole face image into convolutional neural networks (CNNs) or utilize extra facial priors (e. g., facial parsing maps, facial landmarks) to focus on facial structure, thereby maintaining the consistency of the facial structure while restoring facial details.
no code implementations • 31 Aug 2021 • Junjun Jiang, Chenyang Wang, Xianming Liu, Kui Jiang, Jiayi Ma
By this spectral splitting and aggregation strategy (SSAS), we can divide the original hyperspectral image into multiple samples (\emph{from less to more}) to support the efficient training of the network and effectively exploit the spectral correlations among spectrum.
3 code implementations • 27 Jul 2021 • Changan Wang, Qingyu Song, Boshen Zhang, Yabiao Wang, Ying Tai, Xuyi Hu, Chengjie Wang, Jilin Li, Jiayi Ma, Yang Wu
Therefore, we propose a novel count interval partition criterion called Uniform Error Partition (UEP), which always keeps the expected counting error contributions equal for all intervals to minimize the prediction risk.
1 code implementation • 8 Jun 2021 • Jiayi Ma, Yuxin Deng
Modifications on triplet loss that rescale the back-propagated gradients of special pairs have made significant progress on local descriptor learning.
1 code implementation • 31 May 2021 • Pengwei Liang, Junjun Jiang, Xianming Liu, Jiayi Ma
In particular, we estimate the blur amounts of different regions by the internal geometric constraint of the DP data, which measures the defocus disparity between the left and right views.
Ranked #7 on
Image Defocus Deblurring
on DPD (Dual-view)
1 code implementation • 24 May 2021 • JiaMing Wang, Zhenfeng Shao, Xiao Huang, Tao Lu, Ruiqian Zhang, Jiayi Ma
Most existing deep learning-based pan-sharpening methods have several widely recognized issues, such as spectral distortion and insufficient spatial texture enhancement, we propose a novel pan-sharpening convolutional neural network based on a high-pass modification block.
no code implementations • ICCV 2021 • Peng Yi, Zhongyuan Wang, Kui Jiang, Junjun Jiang, Tao Lu, Xin Tian, Jiayi Ma
Most recent video super-resolution (SR) methods either adopt an iterative manner to deal with low-resolution (LR) frames from a temporally sliding window, or leverage the previously estimated SR output to help reconstruct the current frame recurrently.
no code implementations • 30 Jan 2021 • Chengli Peng, Jiayi Ma, Chen Chen, Xiaojie Guo
To verify the efficiency of the proposed bilateral attention decoder, we adopt a lightweight network as the backbone and compare our proposed method with other state-of-the-art real-time semantic segmentation methods on the Cityscapes and Camvid datasets.
1 code implementation • 11 Jan 2021 • Junjun Jiang, Chenyang Wang, Xianming Liu, Jiayi Ma
Second, we elaborate on the facial characteristics and popular datasets used in FSR.
1 code implementation • ICCV 2021 • Zhen Zhong, Guobao Xiao, Linxin Zheng, Yan Lu, Jiayi Ma
We develop a conceptually simple, flexible, and effective framework (named T-Net) for two-view correspondence learning.
1 code implementation • ICCV 2021 • Changan Wang, Qingyu Song, Boshen Zhang, Yabiao Wang, Ying Tai, Xuyi Hu, Chengjie Wang, Jilin Li, Jiayi Ma, Yang Wu
Therefore, we propose a novel count interval partition criterion called Uniform Error Partition (UEP), which always keeps the expected counting error contributions equal for all intervals to minimize the prediction risk.
2 code implementations • 18 May 2020 • Junjun Jiang, He Sun, Xian-Ming Liu, Jiayi Ma
Recently, single gray/RGB image super-resolution reconstruction task has been extensively studied and made significant progress by leveraging the advanced machine learning techniques based on deep convolutional neural networks (DCNNs).
1 code implementation • CVPR 2020 • Yuan Gao, Haoping Bai, Zequn Jie, Jiayi Ma, Kui Jia, Wei Liu
We propose to incorporate neural architecture search (NAS) into general-purpose multi-task learning (GP-MTL).
3 code implementations • CVPR 2020 • Kui Jiang, Zhongyuan Wang, Peng Yi, Chen Chen, Baojin Huang, Yimin Luo, Jiayi Ma, Junjun Jiang
In this work, we explore the multi-scale collaborative representation for rain streaks from the perspective of input image scales and hierarchical deep features in a unified framework, termed multi-scale progressive fusion network (MSPFN) for single image rain streak removal.
Ranked #6 on
Single Image Deraining
on Test2800
1 code implementation • 26 Nov 2019 • Yang Yang, Xiaojie Guo, Jiayi Ma, Lin Ma, Haibin Ling
It is challenging to inpaint face images in the wild, due to the large variation of appearance, such as different poses, expressions and occlusions.
1 code implementation • 24 Nov 2019 • Yuanbin Fu, Jiayi Ma, Lin Ma, Xiaojie Guo
The principle behind is that, for images from multiple domains, the content features can be obtained by a uniform extractor, while (re-)stylization is achieved by mapping the extracted features specifically to different purposes (domains and exemplars).
Generative Adversarial Network
Image-to-Image Translation
+1
1 code implementation • 12 May 2019 • Junjun Jiang, Yi Yu, Zheng Wang, Suhua Tang, Ruimin Hu, Jiayi Ma
In this paper, we present a simple but effective single image SR method based on ensemble learning, which can produce a better performance than that could be obtained from any of SR methods to be ensembled (or called component super-resolvers).
18 code implementations • 28 Feb 2019 • Xiaojie Guo, Siyuan Li, Jinke Yu, Jiawan Zhang, Jiayi Ma, Lin Ma, Wei Liu, Haibin Ling
Being accurate, efficient, and compact is essential to a facial landmark detector for practical use.
1 code implementation • 12 Sep 2018 • Junjun Jiang, Jiayi Ma, Zheng Wang, Chen Chen, Xian-Ming Liu
The key idea of RLPA is to exploit knowledge (e. g., the superpixel based spectral-spatial constraints) from the observed hyperspectral images and apply it to the process of label propagation.
2 code implementations • 3 Sep 2018 • Junjun Jiang, Yi Yu, Suhua Tang, Jiayi Ma, Akiko Aizawa, Kiyoharu Aizawa
To this end, this study incorporates the contextual information of image patch and proposes a powerful and efficient context-patch based face hallucination approach, namely Thresholding Locality-constrained Representation and Reproducing learning (TLcR-RL).
1 code implementation • 28 Jun 2018 • Junjun Jiang, Yi Yu, Jinhui Hu, Suhua Tang, Jiayi Ma
Most of the current face hallucination methods, whether they are shallow learning-based or deep learning-based, all try to learn a relationship model between Low-Resolution (LR) and High-Resolution (HR) spaces with the help of a training set.
1 code implementation • 26 Jun 2018 • Junjun Jiang, Jiayi Ma, Chen Chen, Zhongyuan Wang, Zhihua Cai, Lizhe Wang
(1) Unlike the traditional PCA method based on a whole image, SuperPCA takes into account the diversity in different homogeneous regions, that is, different regions should have different projections.
1 code implementation • CVPR 2019 • Yuan Gao, Jiayi Ma, Mingbo Zhao, Wei Liu, Alan L. Yuille
In this paper, we propose a novel Convolutional Neural Network (CNN) structure for general-purpose multi-task learning (MTL), which enables automatic feature fusing at every layer from different tasks.
Ranked #100 on
Semantic Segmentation
on NYU Depth v2
no code implementations • 12 Sep 2016 • Yuan Gao, Jiayi Ma, Alan L. Yuille
This is based on recent work on sparsity where faces are represented in terms of two dictionaries: a gallery dictionary consisting of one or more examples of each person, and a variation dictionary representing linear nuisance variables (e. g., different lighting conditions, different glasses).
no code implementations • 23 Oct 2014 • Ji Zhao, Deyu Meng, Jiayi Ma
Typically, the region search methods project the score of a classifier into an image plane, and then search the region with the maximal score.
no code implementations • CVPR 2013 • Jiayi Ma, Ji Zhao, Jinwen Tian, Zhuowen Tu, Alan L. Yuille
In the second step, we estimate the transformation using a robust estimator called L 2 E. This is the main novelty of our approach and it enables us to deal with the noise and outliers which arise in the correspondence step.