no code implementations • 7 May 2024 • Bo Yan
However, existing graph condensation methods rely on centralized data storage, which is unfeasible for real-world decentralized data distribution, and overlook data holders' privacy-preserving requirements.
no code implementations • 19 Mar 2024 • Yubin Zheng, Peng Tang, Tianjie Ju, Weidong Qiu, Bo Yan
The intra-client and inter-client consistency learning are introduced to smooth predictions at the data level and avoid confirmation bias of local models.
no code implementations • 31 Jan 2024 • Wenxuan Yang, Weimin Tan, Yuqi Sun, Bo Yan
This paper introduces data-effective learning, aiming to use data in the most impactful way to pre-train foundation models.
1 code implementation • 18 Dec 2023 • Ruian He, Shili Zhou, Yuqi Sun, Ri Cheng, Weimin Tan, Bo Yan
With the rise of real-time rendering and the evolution of display devices, there is a growing demand for post-processing methods that offer high-resolution content in a high frame rate.
no code implementations • 12 Dec 2023 • Ri Cheng, Ruian He, Xuhao Jiang, Shili Zhou, Weimin Tan, Bo Yan
In this paper, we develop a Context-Aware Iteration Policy Network for efficient optical flow estimation, which determines the optimal number of iterations per sample.
no code implementations • 18 Oct 2023 • Bo Yan, Yang Cao, Haoyu Wang, Wenchuan Yang, Junping Du, Chuan Shi
Existing HIN-based recommender systems operate under the assumption of centralized storage and model training.
1 code implementation • 15 Sep 2023 • Ruian He, Zhen Xing, Weimin Tan, Bo Yan
Second, we propose a novel representation diffusion model (RDM) to disentangle 3D latent into facial identity and expression.
no code implementations • 3 Aug 2023 • Shili Zhou, Xuhao Jiang, Weimin Tan, Ruian He, Bo Yan
Motion vectors, one of the compression information, record the motion of the video frames.
no code implementations • 31 Jul 2023 • Ri Cheng, Xuhao Jiang, Ruian He, Shili Zhou, Weimin Tan, Bo Yan
We can use dynamic spatial pruning method to skip redundant computation, but this method cannot properly identify easy regions in VFI tasks without supervision.
no code implementations • 31 Jul 2023 • Shili Zhou, Ruian He, Weimin Tan, Bo Yan
Optical Flow Estimation aims to find the 2D dense motion field between two frames.
1 code implementation • 3 Jul 2023 • Weimin Tan, Siyuan Chen, Bo Yan
Although various few-shot semantic segmentation (FSS) models with different network structures have been proposed, performance improvement has reached a bottleneck.
Ranked #5 on Few-Shot Semantic Segmentation on PASCAL-5i (1-Shot)
no code implementations • 19 Jun 2023 • Yuqi Sun, Ruian He, Weimin Tan, Bo Yan
Given a short speech video, we first build an efficient talking radiance field, and then apply the latest conditional diffusion model for image editing based on the given instructions and guiding implicit representation optimization towards the editing target.
1 code implementation • 24 May 2023 • Yu Ling, Weimin Tan, Bo Yan
Survival analysis aims at modeling the relationship between covariates and event occurrence with some untracked (censored) samples.
no code implementations • 4 May 2023 • Xuhao Jiang, Weimin Tan, Qing Lin, Chenxi Ma, Bo Yan, Liquan Shen
In recent years, many convolutional neural network-based models are designed for JPEG artifacts reduction, and have achieved notable progress.
no code implementations • 26 Apr 2023 • Xuhao Jiang, Weimin Tan, Tian Tan, Bo Yan, Liquan Shen
Image-based single-modality compression learning approaches have demonstrated exceptionally powerful encoding and decoding capabilities in the past few years , but suffer from blur and severe semantics loss at extremely low bitrates.
no code implementations • 2 Apr 2023 • Bo Yan, Cheng Yang, Chuan Shi, Jiawei Liu, Xiaochen Wang
AEHCL designs the intra-event and inter-event contrastive modules to exploit self-supervised AHIN information.
no code implementations • 2 Apr 2023 • Bo Yan, Cheng Yang, Chuan Shi, Yong Fang, Qi Li, Yanfang Ye, Junping Du
In recent years, with the proliferation of graph mining techniques, many researchers investigated these techniques for capturing correlations between cyber entities and achieving high performance.
no code implementations • 21 Nov 2022 • Bo Yan, XingRan Zhao, Yadong Li, Hongbin Wang
And in order to make the most of visual inductive priors, we designed a task-specific inference processing strategy.
2 code implementations • 28 Sep 2022 • Bo Yan, Fengliang Qi, Zhuang Li, Yadong Li, Hongbin Wang
The goal of ACM MMSports2022 DeepSportRadar Instance Segmentation Challenge is to tackle the segmentation of individual humans including players, coaches and referees on a basketball court.
no code implementations • 18 Jul 2022 • Ri Cheng, Yuqi Sun, Bo Yan, Weimin Tan, Chenxi Ma
To address these problems, we propose the MVSRnet, which uses geometry information to extract sharp details from all LR multi-view to support the SR of the LR input view.
no code implementations • 15 Jul 2022 • Xuhao Jiang, Weimin Tan, Ri Cheng, Shili Zhou, Bo Yan
Under stereo settings, the performance of image JPEG artifacts removal can be further improved by exploiting the additional information provided by a second view.
no code implementations • 14 Jul 2022 • Chenxi Ma, Bo Yan, Qing Lin, Weimin Tan, Siming Chen
To enhance the semantic accuracy and the visual quality of the reconstructed image, we explore the multi-modal fusion learning in SISR by proposing a Text-Guided Super-Resolution (TGSR) framework, which can effectively utilize the information from the text and image modalities.
no code implementations • 14 Jul 2022 • Chenxi Ma, Bo Yan, Weimin Tan, Xuhao Jiang
Recent studies of deep learning based stereo image super-resolution (StereoSR) have promoted the development of StereoSR.
no code implementations • 28 Jun 2022 • Bo Yan, Leilei Cao, Zhuang Li, Hongbin Wang
Finally, our approach achieves 63. 008\%AP@0. 50:0. 95 on the test set of CVPR2022 AVA Challenge.
no code implementations • 24 Jun 2022 • Leilei Cao, Zhuang Li, Bo Yan, Feng Zhang, Fengliang Qi, Yuchen Hu, Hongbin Wang
The referring video object segmentation task (RVOS) aims to segment object instances in a given video referred by a language expression in all video frames.
no code implementations • 24 Jun 2022 • Bo Yan, Leilei Cao, Fengliang Qi, Hongbin Wang
Firstly, we designed a context branch based on channel splitting network with transformer to obtain sufficient context information.
no code implementations • CVPR 2022 • Yuqi Sun, Shili Zhou, Ri Cheng, Weimin Tan, Bo Yan, Lang Fu
Specifically, GR stage takes sparse depth map and RGB as input to predict dense depth map by exploiting the correlation between two modals.
no code implementations • 2 Dec 2021 • Fengliang Qi, Bo Yan, Leilei Cao, Hongbin Wang
Person re-identification (re-ID) aims to identify the same person of interest across non-overlapping capturing cameras, which plays an important role in visual surveillance applications and computer vision research areas.
no code implementations • 2 Dec 2021 • Bo Yan, Fengliang Qi, Leilei Cao, Hongbin Wang
Finally, our approach can achieve 40. 2\%AP@0. 50:0. 95 on the test set of ICCV2021 VIPriors instance segmentation challenge.
no code implementations • 2 Dec 2021 • Bo Yan, Leilei Cao, Hongbin Wang
Based on VSPW, we design a Temporal Bilateral Network with Vision Transformer.
1 code implementation • 12 Nov 2021 • Ling Cai, Krzysztof Janowic, Bo Yan, Rui Zhu, Gengchen Mai
Hence, knowledge base completion (KBC) on temporal knowledge bases (TKB), where each statement \textit{may} be associated with a temporal scope, has attracted growing attention.
no code implementations • 7 Nov 2021 • Gengchen Mai, Krzysztof Janowicz, Yingjie Hu, Song Gao, Bo Yan, Rui Zhu, Ling Cai, Ni Lao
A common need for artificial intelligence models in the broader geoscience is to represent and encode various types of spatial data, such as points (e. g., points of interest), polylines (e. g., trajectories), polygons (e. g., administrative regions), graphs (e. g., transportation networks), or rasters (e. g., remote sensing images), in a hidden embedding space so that they can be readily incorporated into deep learning models.
no code implementations • 27 Oct 2021 • Sikai Zhang, Hong Zheng, Hongyi Su, Bo Yan, Jiamou Liu, Song Yang
The main novelty of the model is the integration of time series of four different time granularities: the original time series, together with hourly, daily, and weekly time series.
1 code implementation • 8 Jul 2021 • Ruian He, Zhen Xing, Weimin Tan, Bo Yan
Affective Analysis is not a single task, and the valence-arousal value, expression class, and action unit can be predicted at the same time.
no code implementations • 20 May 2021 • Ran Zhu, Mingkun Yang, Wang Liu, Rujun Song, Bo Yan, Zhuoling Xiao
The technology for Visual Odometry (VO) that estimates the position and orientation of the moving object through analyzing the image sequences captured by on-board cameras, has been well investigated with the rising interest in autonomous driving.
1 code implementation • ICCV 2021 • Jichun Li, Weimin Tan, Bo Yan
Deep learning algorithms have made significant progress in dynamic scene deblurring.
no code implementations • 4 Dec 2020 • Leilei Cao, Tong Yang, Yixu Wang, Bo Yan, Yandong Guo
Thus, our model consists of a pyramid of fully convolutional GANs, wherein the content GAN is responsible for completing contents in the lowest-resolution masked image, and each texture GAN is responsible for synthesizing textures in a higher-resolution image.
no code implementations • 16 Jun 2020 • Matthew Walker, Bo Yan, Yiou Xiao, Yafei Wang, Ayan Acharya
Many tasks that rely on representations of nodes in graphs would benefit if those representations were faithful to distances between nodes in the graph.
no code implementations • 11 Jun 2020 • Ling Cai, Krzysztof Janowicz, Gengchen Mai, Bo Yan, Rui Zhu
In this work we propose a novel deep learning architecture called Traffic Transformer to capture the continuity and periodicity of time series and to model spatial dependency.
Ranked #1 on Time Series Forecasting on US Economy
no code implementations • CVPR 2020 • Bo Yan, Qing Lin, Weimin Tan, Shili Zhou
With the wide use of artistic images, aesthetic quality assessment has been widely concerned.
no code implementations • CVPR 2020 • Bo Yan, Chenxi Ma, Bahetiyaer Bare, Weimin Tan, Steven C. H. Hoi
Under stereo settings, the problems of disparity estimation, stereo magnification and stereo-view synthesis have gathered wide attention.
1 code implementation • 25 Apr 2020 • Gengchen Mai, Krzysztof Janowicz, Ling Cai, Rui Zhu, Blake Regalia, Bo Yan, Meilin Shi, Ni Lao
We also construct a geographic knowledge graph as well as a set of geographic query-answer pairs called DBGeo to evaluate the performance of SE-KGE in comparison to multiple baselines.
2 code implementations • ICLR 2020 • Gengchen Mai, Krzysztof Janowicz, Bo Yan, Rui Zhu, Ling Cai, Ni Lao
The key idea is to use neural networks to convert words in texts to vector space representations based on word positions in a sentence and their contexts, which are suitable for end-to-end training of downstream tasks.
no code implementations • 1 Oct 2019 • Ling Cai, Bo Yan, Gengchen Mai, Krzysztof Janowicz, Rui Zhu
Link prediction is an important and frequently studied task that contributes to an understanding of the structure of knowledge graphs (KGs) in statistical relational learning.
2 code implementations • 30 Sep 2019 • Gengchen Mai, Krzysztof Janowicz, Bo Yan, Rui Zhu, Ling Cai, Ni Lao
Recently, several studies have explored methods for using KG embedding to answer logical queries.
no code implementations • 28 Sep 2019 • Bo Yan, Chuming Lin, Weimin Tan
For video super-resolution, current state-of-the-art approaches either process multiple low-resolution (LR) frames to produce each output high-resolution (HR) frame separately in a sliding window fashion or recurrently exploit the previously estimated HR frames to super-resolve the following frame.
1 code implementation • 9 May 2019 • Weimin Tan, Bo Yan, Chumin Lin, Xuejing Niu
Our idea is built on the reverse mapping from the retargeted images to the given input images.
no code implementations • CVPR 2018 • Weimin Tan, Bo Yan, Bahetiyaer Bare
In this paper, different from image super-resolution (ISR), we propose a novel super-resolution technique called feature super-resolution (FSR), which aims at enhancing the discriminatory power of small size image in order to provide high recognition precision for machine.
no code implementations • 2 May 2018 • Bo Yan, Yiping Liu, Jiamou Liu, Yijin Cai, Hongyi Su, Hong Zheng
We model integration as the interplay between the newcomer and the dynamics network and capture information brokerage using a process of relationship building.