no code implementations • 12 Mar 2024 • Honghao Chen, Xiangxiang Chu, Yongjian Ren, Xin Zhao, Kaiqi Huang
Due to these issues, current CNNs compromise to scale up to 51x51 in the form of stripe convolution (i. e., 51x5 + 5x51) and start to saturate as the kernel size continues growing.
no code implementations • 15 Jan 2024 • Xingzhou Lou, Junge Zhang, Ziyan Wang, Kaiqi Huang, Yali Du
Through the use of pre-trained LMs and the elimination of the need for a ground-truth cost, our method enhances safe policy learning under a diverse set of human-derived free-form natural language constraints.
1 code implementation • 25 Dec 2023 • Xingzhou Lou, Junge Zhang, Timothy J. Norman, Kaiqi Huang, Yali Du
We propose Topology-based multi-Agent Policy gradiEnt (TAPE) for both stochastic and deterministic MAPG methods.
no code implementations • ICCV 2023 • Shiyue Cao, Yueqin Yin, Lianghua Huang, Yu Liu, Xin Zhao, Deli Zhao, Kaiqi Huang
Vector-quantized image modeling has shown great potential in synthesizing high-quality images.
no code implementations • 11 Feb 2023 • Likun Yang, Xiaokun Feng, Xiaotang Chen, Shiyu Zhang, Kaiqi Huang
Dataset analysis illustrates that SpyIn is not only able to support VEIT, but also more challenging compared with other captioning datasets.
1 code implementation • 16 Jan 2023 • Xingzhou Lou, Jiaxian Guo, Junge Zhang, Jun Wang, Kaiqi Huang, Yali Du
We conduct experiments on the Overcooked environment, and evaluate the zero-shot human-AI coordination performance of our method with both behavior-cloned human proxies and real humans.
no code implementations • 5 Jan 2023 • Fei He, Haoyang Zhang, Naiyu Gao, Jian Jia, Yanhu Shan, Xin Zhao, Kaiqi Huang
When using such a pair to predict an object instance on the current frame, not only the generated instance is automatically associated with its precursors on previous frames, but the model gets a good prior for predicting the same object.
no code implementations • 2 Dec 2022 • Jian Jia, Fei He, Naiyu Gao, Xiaotang Chen, Kaiqi Huang
The specificity of the framework lies in a feature disentangle module, which contains learnable semantic queries and a Semantic Spatial Cross-Attention (SSCA) module.
no code implementations • 1 Dec 2022 • Qiyue Yin, Tongtong Yu, Shengqi Shen, Jun Yang, Meijing Zhao, Kaiqi Huang, Bin Liang, Liang Wang
With the breakthrough of AlphaGo, deep reinforcement learning becomes a recognized technique for solving sequential decision-making problems.
no code implementations • 16 Oct 2022 • Yueqin Yin, Lianghua Huang, Yu Liu, Kaiqi Huang
In this work, we first design a group of mechanisms to simulate generative artifacts of popular generators (i. e., GANs, autoregressive models, and diffusion models), given real images.
no code implementations • 22 Jul 2022 • Fei He, Naiyu Gao, Jian Jia, Xin Zhao, Kaiqi Huang
The proposed QueryProp contains two propagation strategies: 1) query propagation is performed from sparse key frames to dense non-key frames to reduce the redundant computation on non-key frames; 2) query propagation is performed from previous key frames to the current key frame to improve feature representation by temporal context modeling.
no code implementations • 2 Jun 2022 • Hao Chen, Guangkai Yang, Junge Zhang, Qiyue Yin, Kaiqi Huang
Specifically, these methods do not explicitly utilize the relationship between agents and cannot adapt to different sizes of inputs.
1 code implementation • CVPR 2022 • Naiyu Gao, Fei He, Jian Jia, Yanhu Shan, Haoyang Zhang, Xin Zhao, Kaiqi Huang
To overcome these limitations, we propose a unified framework for the DPS task by applying a dynamic convolution technique to both the PS and depth prediction tasks.
1 code implementation • 30 May 2022 • Xiaohan Ding, Honghao Chen, Xiangyu Zhang, Kaiqi Huang, Jungong Han, Guiguang Ding
For the extreme simplicity of model structure, we focus on a VGG-style plain model and showcase that such a simple model trained with a RepOptimizer, which is referred to as RepOpt-VGG, performs on par with or better than the recent well-designed models.
no code implementations • 15 Apr 2022 • Shiyu Hu, Xin Zhao, Kaiqi Huang
Single object tracking (SOT) research falls into a cycle -- trackers perform well on most benchmarks but quickly fail in challenging scenarios, causing researchers to doubt the insufficient data content and take more effort to construct larger datasets with more challenging situations.
no code implementations • 2 Mar 2022 • Xiaokun Feng, Xiaotang Chen, Jian Jia, Kaiqi Huang
Sandplay image, as an important psychoanalysis carrier, is a visual scene constructed by the client selecting and placing sand objects (e. g., sand, river, human figures, animals, vegetation, buildings, etc.).
1 code implementation • 26 Feb 2022 • Shiyu Hu, Xin Zhao, Lianghua Huang, Kaiqi Huang
Finally, we design a scientific evaluation procedure using human capabilities as the baseline to judge tracking intelligence.
1 code implementation • 27 Jan 2022 • Qibin Zhou, Dongdong Bai, Junge Zhang, Fuqing Duan, Kaiqi Huang
It is more common in life than perfect-information game.
no code implementations • 15 Nov 2021 • Qiyue Yin, Jun Yang, Kaiqi Huang, Meijing Zhao, Wancheng Ni, Bin Liang, Yan Huang, Shu Wu, Liang Wang
Through this survey, we 1) compare the main difficulties among different kinds of games and the corresponding techniques utilized for achieving professional human level AIs; 2) summarize the mainstream frameworks and techniques that can be properly relied on for developing AIs for complex human-computer gaming; 3) raise the challenges or drawbacks of current techniques in the successful AIs; and 4) try to point out future trends in human-computer gaming AIs.
no code implementations • ICCV 2021 • Jian Jia, Xiaotang Chen, Kaiqi Huang
To fully exploit inter-image relations and aggregate human prior in the model learning process, we construct a Spatial and Semantic Consistency (SSC) framework that consists of two complementary regularizations to achieve spatial and semantic consistency for each attribute.
1 code implementation • 8 Jul 2021 • Jian Jia, Houjing Huang, Xiaotang Chen, Kaiqi Huang
Second, based on the proposed definition, we expose the limitations of the existing datasets, which violate the academic norm and are inconsistent with the essential requirement of practical industry application.
no code implementations • 9 Apr 2021 • Wenzhen Huang, Qiyue Yin, Junge Zhang, Kaiqi Huang
More specifically, we evaluate the effect of an imaginary transition by calculating the change of the loss computed on the real samples when we use the transition to train the action-value and policy functions.
Model-based Reinforcement Learning reinforcement-learning +1
no code implementations • 28 Sep 2020 • Naiyu Gao, Yanhu Shan, Xin Zhao, Kaiqi Huang
Panoptic segmentation (PS) is a complex scene understanding task that requires providing high-quality segmentation for both thing objects and stuff regions.
2 code implementations • 25 May 2020 • Jian Jia, Houjing Huang, Wenjie Yang, Xiaotang Chen, Kaiqi Huang
Despite various methods are proposed to make progress in pedestrian attribute recognition, a crucial problem on existing datasets is often neglected, namely, a large number of identical pedestrian identities in train and test set, which is not consistent with practical application.
Ranked #3 on Pedestrian Attribute Recognition on PETA
1 code implementation • 18 Dec 2019 • Lianghua Huang, Xin Zhao, Kaiqi Huang
Specifically, we propose GlobalTrack, a pure global instance search based tracker that makes no assumption on the temporal consistency of the target's positions and scales.
1 code implementation • International Conference on Computer Vision Workshops 2019 • Dawei Du, Pengfei Zhu, Longyin Wen, Xiao Bian, Haibin Lin, QinGhua Hu, Tao Peng, Jiayu Zheng, Xinyao Wang, Yue Zhang, Liefeng Bo, Hailin Shi, Rui Zhu, Aashish Kumar, Aijin Li, Almaz Zinollayev, Anuar Askergaliyev, Arne Schumann, Binjie Mao, Byeongwon Lee, Chang Liu, Changrui Chen, Chunhong Pan, Chunlei Huo, Da Yu, Dechun Cong, Dening Zeng, Dheeraj Reddy Pailla, Di Li, Dong Wang, Donghyeon Cho, Dongyu Zhang, Furui Bai, George Jose, Guangyu Gao, Guizhong Liu, Haitao Xiong, Hao Qi, Haoran Wang, Heqian Qiu, Hongliang Li, Huchuan Lu, Ildoo Kim, Jaekyum Kim, Jane Shen, Jihoon Lee, Jing Ge, Jingjing Xu, Jingkai Zhou, Jonas Meier, Jun Won Choi, Junhao Hu, Junyi Zhang, Junying Huang, Kaiqi Huang, Keyang Wang, Lars Sommer, Lei Jin, Lei Zhang
Results of 33 object detection algorithms are presented.
1 code implementation • 10 Sep 2019 • Zihao Li, Shu Zhang, Junge Zhang, Kaiqi Huang, Yizhou Wang, Yizhou Yu
In this paper, we propose to incorporate domain knowledge in clinical practice into the model design of universal lesion detectors.
Ranked #8 on Medical Object Detection on DeepLesion
2 code implementations • ICCV 2019 • Naiyu Gao, Yanhu Shan, Yupei Wang, Xin Zhao, Yinan Yu, Ming Yang, Kaiqi Huang
Moreover, incorporating with the learned affinity pyramid, a novel cascaded graph partition module is presented to sequentially generate instances from coarse to fine.
1 code implementation • arXiv:1908.02111 2019 • Huikai Wu, Junge Zhang, Kaiqi Huang
The key idea of the proposed network is to exploit the local similarity of point cloud and the analogy between LR input and HR output.
Ranked #2 on Point Cloud Super Resolution on SHREC15
Graphics Image and Video Processing
1 code implementation • ICCV 2019 • Huikai Wu, Junge Zhang, Kaiqi Huang
In this paper, we aim at automatically searching an efficient network architecture for dense image prediction.
12 code implementations • 28 Mar 2019 • Huikai Wu, Junge Zhang, Kaiqi Huang, Kongming Liang, Yizhou Yu
Modern approaches for semantic segmentation usually employ dilated convolutions in the backbone to extract high-resolution feature maps, which brings heavy computation complexity and memory footprint.
Ranked #40 on Semantic Segmentation on PASCAL Context
no code implementations • 8 Jan 2019 • Xin Zhao, Zhe Liu, Ruolan Hu, Kaiqi Huang
On the other hand, our network obtains the useful features and suppresses the features with less information by a SENet module.
3 code implementations • 29 Dec 2018 • Houjing Huang, Wenjie Yang, Xiaotang Chen, Xin Zhao, Kaiqi Huang, Jinbin Lin, Guan Huang, Dalong Du
Person re-identification (ReID) has achieved significant improvement under the single-domain setting.
2 code implementations • 29 Oct 2018 • Lianghua Huang, Xin Zhao, Kaiqi Huang
(5) Finally, we develop a comprehensive platform for the tracking community that offers full-featured evaluation toolkits, an online evaluation server, and a responsive leaderboard.
no code implementations • CVPR 2018 • Houjing Huang, Dangwei Li, Zhang Zhang, Xiaotang Chen, Kaiqi Huang
Person re-identification (ReID) is the task of retrieving particular persons across different cameras.
1 code implementation • CVPR 2018 • Yan Li, Junge Zhang, Jian-Guo Zhang, Kaiqi Huang
In this work, we retrospect existing methods and demonstrate the necessity to learn discriminative representations for both visual and semantic instances of ZSL.
1 code implementation • CVPR 2018 • Huikai Wu, Shuai Zheng, Junge Zhang, Kaiqi Huang
To address the problem, we present a novel building block for FCNs, namely guided filtering layer, which is designed for efficiently generating a high-resolution output given the corresponding low-resolution one and a high-resolution guidance map.
no code implementations • 27 Feb 2018 • Yan Li, Junge Zhang, Kaiqi Huang, Jian-Guo Zhang
Different from previous MSD methods that directly transfer the pre-trained object detectors from existing categories to new categories, we propose a more reasonable and robust objectness transfer approach for MSD.
no code implementations • 8 Jan 2018 • Yupei Wang, Xin Zhao, Yin Li, Kaiqi Huang
These ConvNet based edge detectors have approached human level performance on standard benchmarks.
no code implementations • CVPR 2017 • Dangwei Li, Xiaotang Chen, Zhang Zhang, Kaiqi Huang
It is a challenging task due to the large variations in person pose, occlusion, background clutter, etc How to extract powerful features is a fundamental problem in ReID and is still an open problem today.
Ranked #110 on Person Re-Identification on Market-1501
2 code implementations • 9 Oct 2017 • Huikai Wu, Yanqi Zong, Junge Zhang, Kaiqi Huang
We also split MSC into training, validation and test set for the convenience of evaluation and comparison.
3 code implementations • CVPR 2018 • Debang Li, Huikai Wu, Junge Zhang, Kaiqi Huang
Image cropping aims at improving the aesthetic quality of images by adjusting their composition.
no code implementations • CVPR 2017 • Yanhua Cheng, Rui Cai, Zhiwei Li, Xin Zhao, Kaiqi Huang
This layer can learn to adjust the contributions of RGB and depth over each pixel for high-performance object recognition.
Ranked #79 on Semantic Segmentation on NYU Depth v2
no code implementations • CVPR 2017 • Yupei Wang, Xin Zhao, Kaiqi Huang
Edge detection had made significant progress with the help of deep Convolutional Networks (ConvNet).
3 code implementations • CVPR 2017 • Weihua Chen, Xiaotang Chen, Jian-Guo Zhang, Kaiqi Huang
In particular, a quadruplet deep network using a margin-based online hard negative mining is proposed based on the quadruplet loss for the person ReID.
2 code implementations • 21 Mar 2017 • Huikai Wu, Shuai Zheng, Junge Zhang, Kaiqi Huang
Concretely, we propose Gaussian-Poisson Equation to formulate the high-resolution image blending problem, which is a joint optimization constrained by the gradient and color information.
Conditional Image Generation Generative Adversarial Network +1
no code implementations • 29 Nov 2016 • Kai Yu, Yang Zhou, Da Li, Zhang Zhang, Kaiqi Huang
Visual surveillance systems have become one of the largest data sources of Big Visual Data in real world.
no code implementations • 17 Nov 2016 • Kai Yu, Biao Leng, Zhang Zhang, Dangwei Li, Kaiqi Huang
Based on GoogLeNet, firstly, a set of mid-level attribute features are discovered by novelly designed detection layers, where a max-pooling based weakly-supervised object detection technique is used to train these layers with only image-level labels without the need of bounding box annotations of pedestrian attributes.
no code implementations • 19 Jul 2016 • Weihua Chen, Xiaotang Chen, Jian-Guo Zhang, Kaiqi Huang
Person re-identification (ReID) focuses on identifying people across different scenes in video surveillance, which is usually formulated as a binary classification task or a ranking task in current person ReID approaches.
no code implementations • CVPR 2016 • Zhang Zhang, Kaiqi Huang, Tieniu Tan, Peipei Yang, Jun Li
For spectral embedding/clustering, it is still an open problem on how to construct an relation graph to reflect the intrinsic structures in data.
no code implementations • 18 Apr 2016 • Yueying Kao, Ran He, Kaiqi Huang
Human beings often assess the aesthetic quality of an image coupled with the identification of the image's semantic content.
Ranked #5 on Aesthetics Quality Assessment on AVA
2 code implementations • 23 Mar 2016 • Dangwei Li, Zhang Zhang, Xiaotang Chen, Haibin Ling, Kaiqi Huang
RAP has in total 41, 585 pedestrian samples, each of which is annotated with 72 attributes as well as viewpoints, occlusions, body parts information.
no code implementations • ICCV 2015 • Yanhua Cheng, Rui Cai, Chi Zhang, Zhiwei Li, Xin Zhao, Kaiqi Huang, Yong Rui
The reasons are in two-fold: (1) existing similarity measures are sensitive to object pose and scale changes, as well as intra-class variations; and (2) effectively fusing RGB and depth cues is still an open problem.
no code implementations • ICCV 2015 • Lianrui Fu, Junge Zhang, Kaiqi Huang
Occlusion is a main challenge for human pose estimation, which is largely ignored in popular tree structure models.
no code implementations • CVPR 2015 • Kangwei Liu, Junge Zhang, Peipei Yang, Kaiqi Huang
al propose the range move algorithms, which are one of the most successful solvers to this problem.
1 code implementation • 12 Feb 2015 • Weihua Chen, Lijun Cao, Xiaotang Chen, Kaiqi Huang
Non-overlapping multi-camera visual object tracking typically consists of two steps: single camera object tracking and inter-camera object tracking.
no code implementations • CVPR 2014 • Kangwei Liu, Junge Zhang, Kaiqi Huang, Tieniu Tan
The MRF energy function is derived from the deformation decomposition model.