Jiahao Cao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jiahao Cao has a total of 19 co-authors. Here are the top co-authors by number of shared publications:
- Yuan Tian
Chongqing Univ. of Science and Technology
2 shared publications - Cong Han
Chongqing University of Science and Technology, Chongqing, China
1 shared publication Guoqing Teng
1 shared publication- Guoqing Teng
School of Computer Science and Engineering, Chongqing University of Science and Technology, Chongqing 400054, China
1 shared publication - Han Wu
School of Computer Science and Engineering, Chongqing University of Science and Technology, Chongqing, China
1 shared publication - Hao Wu
School of Electrical and Electronic Engineering, North China Electric Power University, Beijing 102206, China
1 shared publication
Latest Publications
- Open Access
- 22 April 2026
- Published by PeerJ in PeerJ Computer Science
- Vol. 12, e3793
DTG-LKNet: dual spatio-temporal graphs and large-kernel convolutions network for traffic prediction
Accurate traffic flow prediction is central to Intelligent Transportation Systems yet remains difficult due to non-Euclidean spatial structure, long-range propagation, and time-varying delays. However, existing deep learning methods have key limitations, including reliance on fixed temporal partitioning, small-kernel locality that restricts receptive fields, and graph constructions that lack long-range, dynamic correlations. This work presents dual spatio-temporal graph and large-kernel convolutionnetwork (DTG-LKNet), a Transformer-based architecture that packages a large-kernel convolution module with a dual spatio-temporal graph in a unified framework. On the temporal side, Deformable Patch Sampling learns sampling offsets around salient timestamps, and large kernels expand the effective receptive field without deep dilation stacks. On the spatial side, DTG-LKNet fuses a functional-similarity graph with the physical road network topology to represent long-range correlations. Comprehensive experiments on three large-scale benchmarks demonstrate consistent state-of-the-art performance against strong baselines. This article further visualizes the effective receptive field, validating the effectiveness of the large-kernel design.
- 22 April 2026
- Published by PeerJ in PeerJ Computer Science
- Vol. 12, e3793
Open AccessDTG-LKNet: dual spatio-temporal graphs and large-kernel convolutions network for traffic prediction
Accurate traffic flow prediction is central to Intelligent Transportation Systems yet remains difficult due to non-Euclidean spatial structure, long-range propagation, and time-varying delays. However, existing deep learning methods have key limitations, including reliance on fixed temporal partitioning, small-kernel locality that restricts receptive fields, and graph constructions that lack long-range, dynamic correlations. This work presents dual spatio-temporal graph and large-kernel convolutionnetwork (DTG-LKNet), a Transformer-based architecture that packages a large-kernel convolution module with a dual spatio-temporal graph in a unified framework. On the temporal side, Deformable Patch Sampling learns sampling offsets around salient timestamps, and large kernels expand the effective receptive field without deep dilation stacks. On the spatial side, DTG-LKNet fuses a functional-similarity graph with the physical road network topology to represent long-range correlations. Comprehensive experiments on three large-scale benchmarks demonstrate consistent state-of-the-art performance against strong baselines. This article further visualizes the effective receptive field, validating the effectiveness of the large-kernel design.
- 31 December 2025
- Published by Elsevier in Digital Signal Processing
- Vol. 172
Monocular vehicle pose and shape reconstruction with dual-branch scene-aware and adaptive feature fusion
- 31 December 2025
- Published by Elsevier in Digital Signal Processing
- Vol. 172
Monocular vehicle pose and shape reconstruction with dual-branch scene-aware and adaptive feature fusion
- Open Access
- 21 October 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (20) , 11254
Frequency-Aware and Interactive Spatial-Temporal Graph Convolutional Network for Traffic Flow Prediction
Accurate traffic flow prediction is pivotal for intelligent transportation systems; yet, existing spatial-temporal graph neural networks (STGNNs) struggle to jointly capture the long-term structural stability, short-term dynamics, and multi-scale temporal patterns of road networks. To address these shortcomings, we propose FISTGCN, a Frequency-Aware Interactive Spatial-Temporal Graph Convolutional Network. FISTGCN enriches raw traffic flow features with learnable spatial and temporal embeddings, thereby providing comprehensive spatial-temporal representations for subsequent modeling. Specifically, it utilizes an interactive dynamic graph convolutional block that generates a time-evolving fused adjacency matrix by combining adaptive and dynamic adjacency matrices. It then applies dual sparse graph convolutions with cross-scale interactions to capture multi-scale spatial dependencies. The gated spectral block projects the input features into the frequency domain and adaptively separates low- and high-frequency components using a learnable threshold. It then employs learnable filters to extract features from different frequency bands and adopts a gating mechanism to adaptively fuse low- and high-frequency information, thereby dynamically highlighting short-term fluctuations or long-term trends. Extensive experiments on four benchmark datasets demonstrate that FISTGCN delivers state-of-the-art predictive accuracy while maintaining competitive computational efficiency.
- 21 October 2025
- Published by MDPI AG in Applied Sciences
- Vol. 15 (20) , 11254
Open AccessFrequency-Aware and Interactive Spatial-Temporal Graph Convolutional Network for Traffic Flow Prediction
Accurate traffic flow prediction is pivotal for intelligent transportation systems; yet, existing spatial-temporal graph neural networks (STGNNs) struggle to jointly capture the long-term structural stability, short-term dynamics, and multi-scale temporal patterns of road networks. To address these shortcomings, we propose FISTGCN, a Frequency-Aware Interactive Spatial-Temporal Graph Convolutional Network. FISTGCN enriches raw traffic flow features with learnable spatial and temporal embeddings, thereby providing comprehensive spatial-temporal representations for subsequent modeling. Specifically, it utilizes an interactive dynamic graph convolutional block that generates a time-evolving fused adjacency matrix by combining adaptive and dynamic adjacency matrices. It then applies dual sparse graph convolutions with cross-scale interactions to capture multi-scale spatial dependencies. The gated spectral block projects the input features into the frequency domain and adaptively separates low- and high-frequency components using a learnable threshold. It then employs learnable filters to extract features from different frequency bands and adopts a gating mechanism to adaptively fuse low- and high-frequency information, thereby dynamically highlighting short-term fluctuations or long-term trends. Extensive experiments on four benchmark datasets demonstrate that FISTGCN delivers state-of-the-art predictive accuracy while maintaining competitive computational efficiency.
- Open Access
- 27 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 170-175
CodGRU: Covariate gated attention GRU for short term runoff forecasting
Runoff simulation and forecasting are critical for effective watershed management, yet their complexity and nonlinear characteristics pose significant challenges. The generation of runoff is the result of the interaction between multiple factors such as precipitation, soil moisture, topography, and human activities, making this dynamic process extremely complex and difficult to model accurately. Traditional models are often based on an understanding of these natural processes but may fall short when dealing with complex real-world scenarios. To address these issues, this study has developed a dynamic gated feature selection model with attention mechanisms (CodGRU), specifically designed for multi-watershed hourly runoff forecasting. By dynamically adjusting the weights of input variables and selecting the most important features, CodGRU can better capture the key factors influencing runoff, thereby improving the accuracy of forecasting. The test results of five watersheds on the real dataset show that the CodGRU model performs excellently in multi-watershed hourly-scale forecasting, outperforming the existing mainstream models and providing reliable support for water resource management and disaster forecasting and mitigation.
- 27 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 170-175
Open AccessCodGRU: Covariate gated attention GRU for short term runoff forecasting
Runoff simulation and forecasting are critical for effective watershed management, yet their complexity and nonlinear characteristics pose significant challenges. The generation of runoff is the result of the interaction between multiple factors such as precipitation, soil moisture, topography, and human activities, making this dynamic process extremely complex and difficult to model accurately. Traditional models are often based on an understanding of these natural processes but may fall short when dealing with complex real-world scenarios. To address these issues, this study has developed a dynamic gated feature selection model with attention mechanisms (CodGRU), specifically designed for multi-watershed hourly runoff forecasting. By dynamically adjusting the weights of input variables and selecting the most important features, CodGRU can better capture the key factors influencing runoff, thereby improving the accuracy of forecasting. The test results of five watersheds on the real dataset show that the CodGRU model performs excellently in multi-watershed hourly-scale forecasting, outperforming the existing mainstream models and providing reliable support for water resource management and disaster forecasting and mitigation.