1 code implementation • 14 Mar 2024 • Zhuoxuan Peng, S. -H. Gary Chan
The existing SDG approaches are mainly for image classification and segmentation, and can hardly be extended to our case due to its regression nature and label ambiguity (i. e., ambiguous pixel-level ground truths).
no code implementations • 1 Dec 2023 • Tianlang He, Zhiqiu Xia, Jierun Chen, Haoliang Li, S. -H. Gary Chan
Unsupervised domain adaptation (UDA) seeks to bridge the domain gap between the target and source using unlabeled target data.
1 code implementation • 18 Oct 2023 • Shuhan Zhong, Sizhe Song, Weipeng Zhuo, Guanyao Li, Yang Liu, S. -H. Gary Chan
To handle the multi-scale temporal patterns and multivariate dependencies, we propose a novel temporal patching approach to model the time series as multi-scale patches, and employ MLPs to capture intra- and inter-patch variations and channel-wise correlations.
1 code implementation • 12 Jul 2023 • Weipeng Zhuo, Ka Ho Chiu, Jierun Chen, Ziqi Zhao, S. -H. Gary Chan, Sangtae Ha, Chul-Ho Lee
To build a prediction model to identify the floor number of a new RF signal upon its measurement, conventional approaches using the crowdsourced RF signals assume that at least few labeled signal samples are available on each floor.
2 code implementations • CVPR 2023 • Jierun Chen, Shiu-hong Kao, Hao He, Weipeng Zhuo, Song Wen, Chul-Ho Lee, S. -H. Gary Chan
To achieve faster networks, we revisit popular operators and demonstrate that such low FLOPS is mainly due to frequent memory access of the operators, especially the depthwise convolution.
1 code implementation • CVPR 2022 • Jierun Chen, Tianlang He, Weipeng Zhuo, Li Ma, Sangtae Ha, S. -H. Gary Chan
Extensive experiments on face recognition show that TVConv reduces the computational cost by up to 3. 1x and improves the corresponding throughput by 2. 3x while maintaining a high accuracy compared to the depthwise convolution.
1 code implementation • 30 Dec 2021 • Guanyao Li, Shuhan Zhong, S. -H. Gary Chan, Ruiyuan Li, Chih-Chieh Hung, Wen-Chih Peng
The information fusion module captures the complex spatial-temporal dependency between regions.
1 code implementation • ICCV 2021 • Haoyue Bai, Fengwei Zhou, Lanqing Hong, Nanyang Ye, S. -H. Gary Chan, Zhenguo Li
In this work, we propose robust Neural Architecture Search for OoD generalization (NAS-OoD), which optimizes the architecture with respect to its performance on generated OoD data by gradient descent.
Ranked #1 on Domain Generalization on NICO Vehicle
no code implementations • 20 May 2021 • Haoyue Bai, Song Wen, S. -H. Gary Chan
The classification branch extracts global group priors by learning correlations among image clusters.
no code implementations • 28 Apr 2021 • Haoyue Bai, S. -H. Gary Chan
Noting the scarcity and low quality (in terms of resolution and scene diversity) of the publicly available video crowd datasets, we have collected and built a large-scale video crowd counting datasets, VidCrowd, to contribute to the community.
no code implementations • 12 Jan 2021 • Jierun Chen, Song Wen, S. -H. Gary Chan
In this paper, we propose and study Wild-JDD, a novel learning framework for joint demosaicking and denoising in the wild.
1 code implementation • 31 Dec 2020 • Haoyue Bai, Jiageng Mao, S. -H. Gary Chan
Single image crowd counting is a challenging computer vision problem with wide applications in public safety, city planning, traffic management, etc.
1 code implementation • 17 Dec 2020 • Haoyue Bai, Rui Sun, Lanqing Hong, Fengwei Zhou, Nanyang Ye, Han-Jia Ye, S. -H. Gary Chan, Zhenguo Li
To address that, we propose DecAug, a novel decomposed feature representation and semantic augmentation approach for OoD generalization.
1 code implementation • 9 Sep 2019 • Haoyue Bai, Song Wen, S. -H. Gary Chan
Designing a general crowd counting algorithm applicable to a wide range of crowd images is challenging, mainly due to the possibly large variation in object scales and the presence of many isolated small clusters.
no code implementations • 18 Jan 2019 • Yifeng Zhang, Ka-Ho Chow, S. -H. Gary Chan
In this paper, we develop a Depth-Adaptive Long Short-Term Memory (DA-LSTM) architecture, which can dynamically adjust the structure depending on information distribution without prior knowledge.
no code implementations • 20 Nov 2018 • Ka-Ho Chow, Anish Hiranandani, Yifeng Zhang, S. -H. Gary Chan
Representation learning of pedestrian trajectories transforms variable-length timestamp-coordinate tuples of a trajectory into a fixed-length vector representation that summarizes spatiotemporal characteristics.