Xiaohan Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Xiaohan Zhang has a total of 60 co-authors. Here are the top co-authors by number of shared publications:
- Hui-Liang Shen
Zhejiang University
11 shared publications - Xiaokai Bai
Zhejiang University
8 shared publications - Si-Yuan Cao
Zhejiang University
6 shared publications - Zhe Wu
College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China
6 shared publications - Xue Zhang
College of Information Science and Electronic Engineering, Zhejiang University, Hangzhou, China
5 shared publications - Beinan Yu
College of Computer Science and Technology, Zhejiang University, Hangzhou, China
4 shared publications
Latest Publications
- 3 October 2026
- Published by Springer Nature
- p. 416-434
Recurrent Cross-View Object Geo-Localization
- 17 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. PP (15209210) , 1-13
Boosting Instance Awareness Via Cross-View Correlation With 4D Radar and Camera for 3D Object Detection
4D millimeter-wave radar has emerged as a promising sensing modality for autonomous driving due to its robustness and affordability. However, its sparse and weak geometric cues make reliable instance activation difficult, limiting the effectiveness of existing radar–camera fusion paradigms. BEV-level fusion offers global scene understanding but suffers from weak instance focus, while perspective-level fusion captures instance details but lacks holistic context. To address these limitations, we propose SIFormer, a scene–instance aware transformer for 3D object detection using 4D radar and camera. SIFormer first suppresses background noise during view transformation through segmentation- and depth-guided localization. It then introduces a cross-view activation mechanism that injects 2D instance cues into BEV space, enabling reliable instance awareness under weak radar geometry. Finally, a transformer-based fusion module aggregates complementary image semantics and radar geometry for robust perception. As a result, with the aim of enhancing instance awareness, SIFormer bridges the gap between the two paradigms, combining their complementary strengths to address inherent sparse nature of radar and improve detection accuracy. Experiments demonstrate that SIFormer achieves state-of-the-art performance on View-of-Delft, TJ4DRadSet and NuScenes datasets. The code will be made publicly available.
- 17 April 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. PP (15209210) , 1-13
Boosting Instance Awareness Via Cross-View Correlation With 4D Radar and Camera for 3D Object Detection
4D millimeter-wave radar has emerged as a promising sensing modality for autonomous driving due to its robustness and affordability. However, its sparse and weak geometric cues make reliable instance activation difficult, limiting the effectiveness of existing radar–camera fusion paradigms. BEV-level fusion offers global scene understanding but suffers from weak instance focus, while perspective-level fusion captures instance details but lacks holistic context. To address these limitations, we propose SIFormer, a scene–instance aware transformer for 3D object detection using 4D radar and camera. SIFormer first suppresses background noise during view transformation through segmentation- and depth-guided localization. It then introduces a cross-view activation mechanism that injects 2D instance cues into BEV space, enabling reliable instance awareness under weak radar geometry. Finally, a transformer-based fusion module aggregates complementary image semantics and radar geometry for robust perception. As a result, with the aim of enhancing instance awareness, SIFormer bridges the gap between the two paradigms, combining their complementary strengths to address inherent sparse nature of radar and improve detection accuracy. Experiments demonstrate that SIFormer achieves state-of-the-art performance on View-of-Delft, TJ4DRadSet and NuScenes datasets. The code will be made publicly available.
- Open Access
Boosting Instance Awareness via Cross-View Correlation with 4D Radar and Camera for 3D Object Detection
4D millimeter-wave radar has emerged as a promising sensing modality for autonomous driving due to its robustness and affordability. However, its sparse and weak geometric cues make reliable instance activation difficult, limiting the effectiveness of existing radar-camera fusion paradigms. BEV-level fusion offers global scene understanding but suffers from weak instance focus, while perspective-level fusion captures instance details but lacks holistic context. To address these limitations, we propose SIFormer, a scene-instance aware transformer for 3D object detection using 4D radar and camera. SIFormer first suppresses background noise during view transformation through segmentation- and depth-guided localization. It then introduces a cross-view activation mechanism that injects 2D instance cues into BEV space, enabling reliable instance awareness under weak radar geometry. Finally, a transformer-based fusion module aggregates complementary image semantics and radar geometry for robust perception. As a result, with the aim of enhancing instance awareness, SIFormer bridges the gap between the two paradigms, combining their complementary strengths to address inherent sparse nature of radar and improve detection accuracy. Experiments demonstrate that SIFormer achieves state-of-the-art performance on View-of-Delft, TJ4DRadSet and NuScenes datasets. Source code is available at github.com/shawnnnkb/SIFormer.
Open AccessBoosting Instance Awareness via Cross-View Correlation with 4D Radar and Camera for 3D Object Detection
4D millimeter-wave radar has emerged as a promising sensing modality for autonomous driving due to its robustness and affordability. However, its sparse and weak geometric cues make reliable instance activation difficult, limiting the effectiveness of existing radar-camera fusion paradigms. BEV-level fusion offers global scene understanding but suffers from weak instance focus, while perspective-level fusion captures instance details but lacks holistic context. To address these limitations, we propose SIFormer, a scene-instance aware transformer for 3D object detection using 4D radar and camera. SIFormer first suppresses background noise during view transformation through segmentation- and depth-guided localization. It then introduces a cross-view activation mechanism that injects 2D instance cues into BEV space, enabling reliable instance awareness under weak radar geometry. Finally, a transformer-based fusion module aggregates complementary image semantics and radar geometry for robust perception. As a result, with the aim of enhancing instance awareness, SIFormer bridges the gap between the two paradigms, combining their complementary strengths to address inherent sparse nature of radar and improve detection accuracy. Experiments demonstrate that SIFormer achieves state-of-the-art performance on View-of-Delft, TJ4DRadSet and NuScenes datasets. Source code is available at github.com/shawnnnkb/SIFormer.
- Open Access
SD4R: Sparse-to-Dense Learning for 3D Object Detection with 4D Radar
4D radar measurements offer an affordable and weather-robust solution for 3D perception. However, the inherent sparsity and noise of radar point clouds present significant challenges for accurate 3D object detection, underscoring the need for effective and robust point clouds densification. Despite recent progress, existing densification methods often fail to address the extreme sparsity of 4D radar point clouds and exhibit limited robustness when processing scenes with a small number of points. In this paper, we propose SD4R, a novel framework that transforms sparse radar point clouds into dense representations. SD4R begins by utilizing a foreground point generator (FPG) to mitigate noise propagation and produce densified point clouds. Subsequently, a logit-query encoder (LQE) enhances conventional pillarization, resulting in robust feature representations. Through these innovations, our SD4R demonstrates strong capability in both noise reduction and foreground point densification. Extensive experiments conducted on the publicly available View-of-Delft dataset demonstrate that SD4R achieves state-of-the-art performance. Source code is available at https://github.com/lancelot0805/SD4R.
Open AccessSD4R: Sparse-to-Dense Learning for 3D Object Detection with 4D Radar
4D radar measurements offer an affordable and weather-robust solution for 3D perception. However, the inherent sparsity and noise of radar point clouds present significant challenges for accurate 3D object detection, underscoring the need for effective and robust point clouds densification. Despite recent progress, existing densification methods often fail to address the extreme sparsity of 4D radar point clouds and exhibit limited robustness when processing scenes with a small number of points. In this paper, we propose SD4R, a novel framework that transforms sparse radar point clouds into dense representations. SD4R begins by utilizing a foreground point generator (FPG) to mitigate noise propagation and produce densified point clouds. Subsequently, a logit-query encoder (LQE) enhances conventional pillarization, resulting in robust feature representations. Through these innovations, our SD4R demonstrates strong capability in both noise reduction and foreground point densification. Extensive experiments conducted on the publicly available View-of-Delft dataset demonstrate that SD4R achieves state-of-the-art performance. Source code is available at https://github.com/lancelot0805/SD4R.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (1) , 266-278
Remote PPG Measurement Using a Synergistic Time-Frequency Network
Remote photoplethysmography (rPPG) aims to estimate the blood volume pulse (BVP) signal from facial videos. Existing rPPG approaches still suffer from limitations. We attribute this issue to two primary problems: (1) the reliance solely on time-domain processing that makes the signal susceptible to interference, and (2) the presence of a phase discrepancy between the supervision signal and the ground-truth PPG. To address these problems, we propose TFSNet, a novel time-frequency synergy network for rPPG signal estimation and heart rate prediction. Specifically, we leverage time-frequency fusion (TFF) module, which integrates frequency-domain information into the learning process to enrich the feature representations. Additionally, we introduce the amplitude-phase decoupling (APD) module, which apply phase compensation in frequency domain to mitigate the adverse effects of incorrect phase supervision. Extensive experiments demonstrate that TFSNet achieves state-of-the-art performance, significantly outperforming current approaches in both accuracy and robustness.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Journal of Biomedical and Health Informatics
- Vol. 30 (1) , 266-278
Remote PPG Measurement Using a Synergistic Time-Frequency Network
Remote photoplethysmography (rPPG) aims to estimate the blood volume pulse (BVP) signal from facial videos. Existing rPPG approaches still suffer from limitations. We attribute this issue to two primary problems: (1) the reliance solely on time-domain processing that makes the signal susceptible to interference, and (2) the presence of a phase discrepancy between the supervision signal and the ground-truth PPG. To address these problems, we propose TFSNet, a novel time-frequency synergy network for rPPG signal estimation and heart rate prediction. Specifically, we leverage time-frequency fusion (TFF) module, which integrates frequency-domain information into the learning process to enrich the feature representations. Additionally, we introduce the amplitude-phase decoupling (APD) module, which apply phase compensation in frequency domain to mitigate the adverse effects of incorrect phase supervision. Extensive experiments demonstrate that TFSNet achieves state-of-the-art performance, significantly outperforming current approaches in both accuracy and robustness.
- 28 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-16
GeoFormer: Boosting Object Distinguishing and Prompt Understanding for Cross-View Object Geo-Localization
Cross-view object geo-localization (CVOGL) determines the geographic location of an object on the satellite view reference image. The object is indicated by a point prompt in a ground- or drone-view query image. Despite wide applications, the current CVOGL method still suffers from limited performance, and the reason for this limitation remains unknown. In this work, we analyze CVOGL and find two primary challenges that hinder its performance, i.e., 1) point prompt understanding and 2) similar-appearance object distinguishing. Therefore, we propose a novel end-to-end framework called object geo-localization transformer (GeoFormer). Specifically, we leverage the knowledge of the segment anything model (SAM) through two settings for accurate point prompt understanding, i.e., using SAM during training and inference (GeoFormer) or using SAM only in training via knowledge distillation (GeoFormer-KD). Additionally, we devise an information aggregation module (IAM) to leverage local and global perception for similar-appearance object distinguishing. Except for the public dataset, we manually annotated a new dataset that contains 1642 image pairs for further comparison. Experiments show that our method significantly outperforms the previous work. Notably, the tiny versions of our method (GeoFormer-t and GeoFormer-t-KD) maintain state-of-the-art performance while substantially reducing parameter costs. Our code and the new dataset will be made available at https://github.com/Temperature-ai/GeoFormer.
- 28 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-16
GeoFormer: Boosting Object Distinguishing and Prompt Understanding for Cross-View Object Geo-Localization
Cross-view object geo-localization (CVOGL) determines the geographic location of an object on the satellite view reference image. The object is indicated by a point prompt in a ground- or drone-view query image. Despite wide applications, the current CVOGL method still suffers from limited performance, and the reason for this limitation remains unknown. In this work, we analyze CVOGL and find two primary challenges that hinder its performance, i.e., 1) point prompt understanding and 2) similar-appearance object distinguishing. Therefore, we propose a novel end-to-end framework called object geo-localization transformer (GeoFormer). Specifically, we leverage the knowledge of the segment anything model (SAM) through two settings for accurate point prompt understanding, i.e., using SAM during training and inference (GeoFormer) or using SAM only in training via knowledge distillation (GeoFormer-KD). Additionally, we devise an information aggregation module (IAM) to leverage local and global perception for similar-appearance object distinguishing. Except for the public dataset, we manually annotated a new dataset that contains 1642 image pairs for further comparison. Experiments show that our method significantly outperforms the previous work. Notably, the tiny versions of our method (GeoFormer-t and GeoFormer-t-KD) maintain state-of-the-art performance while substantially reducing parameter costs. Our code and the new dataset will be made available at https://github.com/Temperature-ai/GeoFormer.
- 24 October 2025
- Published by Elsevier in Pattern Recognition
- Vol. 172
SADet: A semantic-aware tiny object detection network against missed detection
- 24 October 2025
- Published by Elsevier in Pattern Recognition
- Vol. 172
SADet: A semantic-aware tiny object detection network against missed detection
- 18 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-14
TEFormer: Thermal Infrared Image Enhancement by Preserving Spatial Consistency and Details
Thermal infrared (TIR) images suffer from low contrast due to the atmospheric thermal radiation effect, especially under extreme conditions like low temperature. TIR image enhancement aims to improve image contrast, but previous enhancement approaches usually produce enhanced results with two limitations: spatial inconsistency and detail blurring. To deal with the limitations, we propose a novel TIR image enhancement method, named TEFormer, to preserve spatial consistency and restore fine-grained details during image enhancement. To preserve spatial consistency, we devise the global enhancement module (GEM) to enhance the low-resolution representation. The GEM performs long-range interactions across spatial dimensions and channel dimensions to condition the enhancement curve fitting. To keep details clear, we design the local enhancement module (LEM) as the decoding unit. The LEM injects additional detail structures into the enhanced low-resolution representation for high-resolution reconstruction. Besides, we further apply histogram-based supervision to facilitate learning in intensity distribution of clear images. Extensive experimental results on three challenging benchmarks demonstrate that the proposed method outperforms other state-of-the-art approaches.
- 18 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 63 (01962892) , 1-14
TEFormer: Thermal Infrared Image Enhancement by Preserving Spatial Consistency and Details
Thermal infrared (TIR) images suffer from low contrast due to the atmospheric thermal radiation effect, especially under extreme conditions like low temperature. TIR image enhancement aims to improve image contrast, but previous enhancement approaches usually produce enhanced results with two limitations: spatial inconsistency and detail blurring. To deal with the limitations, we propose a novel TIR image enhancement method, named TEFormer, to preserve spatial consistency and restore fine-grained details during image enhancement. To preserve spatial consistency, we devise the global enhancement module (GEM) to enhance the low-resolution representation. The GEM performs long-range interactions across spatial dimensions and channel dimensions to condition the enhancement curve fitting. To keep details clear, we design the local enhancement module (LEM) as the decoding unit. The LEM injects additional detail structures into the enhanced low-resolution representation for high-resolution reconstruction. Besides, we further apply histogram-based supervision to facilitate learning in intensity distribution of clear images. Extensive experimental results on three challenging benchmarks demonstrate that the proposed method outperforms other state-of-the-art approaches.
- Open Access
Recurrent Cross-View Object Geo-Localization
Cross-view object geo-localization (CVOGL) aims to determine the location of a specific object in high-resolution satellite imagery given a query image with a point prompt. Existing approaches treat CVOGL as a one-shot detection task, directly regressing object locations from cross-view information aggregation, but they are vulnerable to feature noise and lack mechanisms for error correction. In this paper, we propose ReCOT, a Recurrent Cross-view Object geo-localization Transformer, which reformulates CVOGL as a recurrent localization task. ReCOT introduces a set of learnable tokens that encode task-specific intent from the query image and prompt embeddings, and iteratively attend to the reference features to refine the predicted location. To enhance this recurrent process, we incorporate two complementary modules: (1) a SAM-based knowledge distillation strategy that transfers segmentation priors from the Segment Anything Model (SAM) to provide clearer semantic guidance without additional inference cost, and (2) a Reference Feature Enhancement Module (RFEM) that introduces a hierarchical attention to emphasize object-relevant regions in the reference features. Extensive experiments on standard CVOGL benchmarks demonstrate that ReCOT achieves state-of-the-art (SOTA) performance while reducing parameters by 60% compared to previous SOTA approaches.
Open AccessRecurrent Cross-View Object Geo-Localization
Cross-view object geo-localization (CVOGL) aims to determine the location of a specific object in high-resolution satellite imagery given a query image with a point prompt. Existing approaches treat CVOGL as a one-shot detection task, directly regressing object locations from cross-view information aggregation, but they are vulnerable to feature noise and lack mechanisms for error correction. In this paper, we propose ReCOT, a Recurrent Cross-view Object geo-localization Transformer, which reformulates CVOGL as a recurrent localization task. ReCOT introduces a set of learnable tokens that encode task-specific intent from the query image and prompt embeddings, and iteratively attend to the reference features to refine the predicted location. To enhance this recurrent process, we incorporate two complementary modules: (1) a SAM-based knowledge distillation strategy that transfers segmentation priors from the Segment Anything Model (SAM) to provide clearer semantic guidance without additional inference cost, and (2) a Reference Feature Enhancement Module (RFEM) that introduces a hierarchical attention to emphasize object-relevant regions in the reference features. Extensive experiments on standard CVOGL benchmarks demonstrate that ReCOT achieves state-of-the-art (SOTA) performance while reducing parameters by 60% compared to previous SOTA approaches.
- 28 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-13
Remote Heart Rate Measurement Using UWB Radar via Multistage Adaptive Estimation
Radar technology is widely recognized for its ability to monitor biological signals without contact. However, heart rate monitoring accuracy is often compromised by interference from respiratory harmonics, especially when their frequencies overlap with heartbeat signals. To solve this challenge, we propose an Adaptive Heart Rate Measurement method using impulse-radio ultra-wideband radar. This method employs a multistage process to enhance heart rate estimation. First, we selectively intercept phase information within the effective range and apply weighting factors to suppress the multipath interference effect. Second, we dynamically adjust notch filter bandwidth based on respiratory asymmetry and pause duration between breaths, preventing excessive suppression of frequencies near respiratory harmonics. Third, we separate the heartbeat and its second harmonic to examine their correspondence and use historical data as prior information for robust heart rate estimation. Experimental results on real-world data demonstrate that the proposed method outperforms existing non-contact heart rate measurement techniques.
- 28 March 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 74 (00189456) , 1-13
Remote Heart Rate Measurement Using UWB Radar via Multistage Adaptive Estimation
Radar technology is widely recognized for its ability to monitor biological signals without contact. However, heart rate monitoring accuracy is often compromised by interference from respiratory harmonics, especially when their frequencies overlap with heartbeat signals. To solve this challenge, we propose an Adaptive Heart Rate Measurement method using impulse-radio ultra-wideband radar. This method employs a multistage process to enhance heart rate estimation. First, we selectively intercept phase information within the effective range and apply weighting factors to suppress the multipath interference effect. Second, we dynamically adjust notch filter bandwidth based on respiratory asymmetry and pause duration between breaths, preventing excessive suppression of frequencies near respiratory harmonics. Third, we separate the heartbeat and its second harmonic to examine their correspondence and use historical data as prior information for robust heart rate estimation. Experimental results on real-world data demonstrate that the proposed method outperforms existing non-contact heart rate measurement techniques.