Qingfan Wang
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
Qingfan Wang has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Dongyang Xu
Tsinghua University
3 shared publications Lei Chen
SenseTime Research
3 shared publications- Bingbing Nie
School of Vehicle and Mobility, Tsinghua University
2 shared publications - Ji Ma
Peking University
2 shared publications Xiangyun Zeng
SenseTime Research
2 shared publications- Chen Lv
School of Mechanical and Aerospace Engineering Nanyang Technological University Singapore Singapore
1 shared publication
Latest Publications
- 12 May 2025
- Published by Springer Nature
- p. 266-277
AHMF: Adaptive Hybrid-Memory-Fusion Model for Driver Attention Prediction
- 12 May 2025
- Published by Springer Nature
- p. 266-277
AHMF: Adaptive Hybrid-Memory-Fusion Model for Driver Attention Prediction
- Open Access
AHMF: Adaptive Hybrid-Memory-Fusion Model for Driver Attention Prediction
Accurate driver attention prediction can serve as a critical reference for intelligent vehicles in understanding traffic scenes and making informed driving decisions. Though existing studies on driver attention prediction improved performance by incorporating advanced saliency detection techniques, they overlooked the opportunity to achieve human-inspired prediction by analyzing driving tasks from a cognitive science perspective. During driving, drivers' working memory and long-term memory play crucial roles in scene comprehension and experience retrieval, respectively. Together, they form situational awareness, facilitating drivers to quickly understand the current traffic situation and make optimal decisions based on past driving experiences. To explicitly integrate these two types of memory, this paper proposes an Adaptive Hybrid-Memory-Fusion (AHMF) driver attention prediction model to achieve more human-like predictions. Specifically, the model first encodes information about specific hazardous stimuli in the current scene to form working memories. Then, it adaptively retrieves similar situational experiences from the long-term memory for final prediction. Utilizing domain adaptation techniques, the model performs parallel training across multiple datasets, thereby enriching the accumulated driving experience within the long-term memory module. Compared to existing models, our model demonstrates significant improvements across various metrics on multiple public datasets, proving the effectiveness of integrating hybrid memories in driver attention prediction.
Open AccessAHMF: Adaptive Hybrid-Memory-Fusion Model for Driver Attention Prediction
Accurate driver attention prediction can serve as a critical reference for intelligent vehicles in understanding traffic scenes and making informed driving decisions. Though existing studies on driver attention prediction improved performance by incorporating advanced saliency detection techniques, they overlooked the opportunity to achieve human-inspired prediction by analyzing driving tasks from a cognitive science perspective. During driving, drivers' working memory and long-term memory play crucial roles in scene comprehension and experience retrieval, respectively. Together, they form situational awareness, facilitating drivers to quickly understand the current traffic situation and make optimal decisions based on past driving experiences. To explicitly integrate these two types of memory, this paper proposes an Adaptive Hybrid-Memory-Fusion (AHMF) driver attention prediction model to achieve more human-like predictions. Specifically, the model first encodes information about specific hazardous stimuli in the current scene to form working memories. Then, it adaptively retrieves similar situational experiences from the long-term memory for final prediction. Utilizing domain adaptation techniques, the model performs parallel training across multiple datasets, thereby enriching the accumulated driving experience within the long-term memory module. Compared to existing models, our model demonstrates significant improvements across various metrics on multiple public datasets, proving the effectiveness of integrating hybrid memories in driver attention prediction.
- Open Access
Risk-Aware Vehicle Trajectory Prediction Under Safety-Critical Scenarios
Trajectory prediction is significant for intelligent vehicles to achieve high-level autonomous driving, and a lot of relevant research achievements have been made recently. Despite the rapid development, most existing studies solely focused on normal safe scenarios while largely neglecting safety-critical scenarios, particularly those involving imminent collisions. This oversight may result in autonomous vehicles lacking the essential predictive ability in such situations, posing a significant threat to safety. To tackle these, this paper proposes a risk-aware trajectory prediction framework tailored to safety-critical scenarios. Leveraging distinctive hazardous features, we develop three core risk-aware components. First, we introduce a risk-incorporated scene encoder, which augments conventional encoders with quantitative risk information to achieve risk-aware encoding of hazardous scene contexts. Next, we incorporate endpoint-risk-combined intention queries as prediction priors in the decoder to ensure that the predicted multimodal trajectories cover both various spatial intentions and risk levels. Lastly, an auxiliary risk prediction task is implemented for the ultimate risk-aware prediction. Furthermore, to support model training and performance evaluation, we introduce a safety-critical trajectory prediction dataset and tailored evaluation metrics. We conduct comprehensive evaluations and compare our model with several SOTA models. Results demonstrate the superior performance of our model, with a significant improvement in most metrics. This prediction advancement enables autonomous vehicles to execute correct collision avoidance maneuvers under safety-critical scenarios, eventually enhancing road traffic safety.
Open AccessRisk-Aware Vehicle Trajectory Prediction Under Safety-Critical Scenarios
Trajectory prediction is significant for intelligent vehicles to achieve high-level autonomous driving, and a lot of relevant research achievements have been made recently. Despite the rapid development, most existing studies solely focused on normal safe scenarios while largely neglecting safety-critical scenarios, particularly those involving imminent collisions. This oversight may result in autonomous vehicles lacking the essential predictive ability in such situations, posing a significant threat to safety. To tackle these, this paper proposes a risk-aware trajectory prediction framework tailored to safety-critical scenarios. Leveraging distinctive hazardous features, we develop three core risk-aware components. First, we introduce a risk-incorporated scene encoder, which augments conventional encoders with quantitative risk information to achieve risk-aware encoding of hazardous scene contexts. Next, we incorporate endpoint-risk-combined intention queries as prediction priors in the decoder to ensure that the predicted multimodal trajectories cover both various spatial intentions and risk levels. Lastly, an auxiliary risk prediction task is implemented for the ultimate risk-aware prediction. Furthermore, to support model training and performance evaluation, we introduce a safety-critical trajectory prediction dataset and tailored evaluation metrics. We conduct comprehensive evaluations and compare our model with several SOTA models. Results demonstrate the superior performance of our model, with a significant improvement in most metrics. This prediction advancement enables autonomous vehicles to execute correct collision avoidance maneuvers under safety-critical scenarios, eventually enhancing road traffic safety.
- Open Access
M2DA: Multi-Modal Fusion Transformer Incorporating Driver Attention for Autonomous Driving
End-to-end autonomous driving has witnessed remarkable progress. However, the extensive deployment of autonomous vehicles has yet to be realized, primarily due to 1) inefficient multi-modal environment perception: how to integrate data from multi-modal sensors more efficiently; 2) non-human-like scene understanding: how to effectively locate and predict critical risky agents in traffic scenarios like an experienced driver. To overcome these challenges, in this paper, we propose a Multi-Modal fusion transformer incorporating Driver Attention (M2DA) for autonomous driving. To better fuse multi-modal data and achieve higher alignment between different modalities, a novel Lidar-Vision-Attention-based Fusion (LVAFusion) module is proposed. By incorporating driver attention, we empower the human-like scene understanding ability to autonomous vehicles to identify crucial areas within complex scenarios precisely and ensure safety. We conduct experiments on the CARLA simulator and achieve state-of-the-art performance with less data in closed-loop benchmarks. Source codes are available at https://anonymous.4open.science/r/M2DA-4772.
Open AccessM2DA: Multi-Modal Fusion Transformer Incorporating Driver Attention for Autonomous Driving
End-to-end autonomous driving has witnessed remarkable progress. However, the extensive deployment of autonomous vehicles has yet to be realized, primarily due to 1) inefficient multi-modal environment perception: how to integrate data from multi-modal sensors more efficiently; 2) non-human-like scene understanding: how to effectively locate and predict critical risky agents in traffic scenarios like an experienced driver. To overcome these challenges, in this paper, we propose a Multi-Modal fusion transformer incorporating Driver Attention (M2DA) for autonomous driving. To better fuse multi-modal data and achieve higher alignment between different modalities, a novel Lidar-Vision-Attention-based Fusion (LVAFusion) module is proposed. By incorporating driver attention, we empower the human-like scene understanding ability to autonomous vehicles to identify crucial areas within complex scenarios precisely and ensure safety. We conduct experiments on the CARLA simulator and achieve state-of-the-art performance with less data in closed-loop benchmarks. Source codes are available at https://anonymous.4open.science/r/M2DA-4772.
- 29 January 2024
- Published by ASME International in Journal of Biomechanical Engineering
- Vol. 146 (3) , 1-48
A Lightweight Pre-Crash Occupant Injury Prediction Model Distills Knowledge From Its Post-Crash Counterpart
Accurate occupant injury prediction in near-collision scenarios is vital in guiding intelligent vehicles to find the optimal collision condition with minimal injury risks. Existing studies focused on boosting prediction performance by introducing deep-learning models but encountered computational burdens due to the inherent high model complexity. To better balance these two traditionally contradictory factors, this study proposed a training method for pre-crash injury prediction models, namely, knowledge distillation (KD)-based training. This method was inspired by the idea of knowledge distillation, an emerging model compression method. Technically, we first trained a high-accuracy injury prediction model using informative post-crash sequence inputs (i.e., vehicle crash pulses) and a relatively complex network architecture as an experienced “teacher”. Following this, a lightweight pre-crash injury prediction model (“student”) learned both from the ground truth in output layers (i.e., conventional prediction loss) and its teacher in intermediate layers (i.e., distillation loss). In such a step-by-step teaching framework, the pre-crash model significantly improved the prediction accuracy of occupant's head abbreviated injury scale (AIS) (i.e., from 77.2% to 83.2%) without sacrificing computational efficiency. Multiple validation experiments proved the effectiveness of the proposed KD-based training framework. This study is expected to provide reference to balancing prediction accuracy and computational efficiency of pre-crash injury prediction models, promoting the further safety improvement of next-generation intelligent vehicles.
- 29 January 2024
- Published by ASME International in Journal of Biomechanical Engineering
- Vol. 146 (3) , 1-48
A Lightweight Pre-Crash Occupant Injury Prediction Model Distills Knowledge From Its Post-Crash Counterpart
Accurate occupant injury prediction in near-collision scenarios is vital in guiding intelligent vehicles to find the optimal collision condition with minimal injury risks. Existing studies focused on boosting prediction performance by introducing deep-learning models but encountered computational burdens due to the inherent high model complexity. To better balance these two traditionally contradictory factors, this study proposed a training method for pre-crash injury prediction models, namely, knowledge distillation (KD)-based training. This method was inspired by the idea of knowledge distillation, an emerging model compression method. Technically, we first trained a high-accuracy injury prediction model using informative post-crash sequence inputs (i.e., vehicle crash pulses) and a relatively complex network architecture as an experienced “teacher”. Following this, a lightweight pre-crash injury prediction model (“student”) learned both from the ground truth in output layers (i.e., conventional prediction loss) and its teacher in intermediate layers (i.e., distillation loss). In such a step-by-step teaching framework, the pre-crash model significantly improved the prediction accuracy of occupant's head abbreviated injury scale (AIS) (i.e., from 77.2% to 83.2%) without sacrificing computational efficiency. Multiple validation experiments proved the effectiveness of the proposed KD-based training framework. This study is expected to provide reference to balancing prediction accuracy and computational efficiency of pre-crash injury prediction models, promoting the further safety improvement of next-generation intelligent vehicles.
- 5 January 2024
- Published by Springer Nature in Journal of Thrombosis and Thrombolysis
- Vol. 57 (3) , 390-401
Effects of elevated remnant cholesterol on outcomes of acute ischemic stroke patients receiving mechanical thrombectomy
Objective: Large cohort studies provided evidence that elevated remnant cholesterol (RC) was an important risk factor for ischemic stroke. However, the association between high RC and clinical outcomes in acute ischemic stroke (AIS) individuals was still undetermined. Methods: This retrospective study enrolled 165 AIS patients undergoing mechanical thrombectomy in one tertiary stroke center. We divided patients into two groups based on the median of their RC levels (0.49 mmol/L). The modified Rankin Scale (mRS) was used to evaluate the primary outcome 90 days after the onset of symptoms. The mRS scores ≤ 2 and ≤ 1 at 90 days were deemed as favorable and excellent outcomes, respectively. Results: In the overall AIS patients undergoing mechanical thrombectomy, there was no obvious distinction between the high and low RC group at 90-day favorable outcome (41.0% vs. 47.1%, P = 0.431) or excellent outcome (23.1% vs. 31.0%, P = 0.252). In the subgroup analysis stratified by stroke etiology, non-large artery atherosclerosis (non-LAA) stroke patients yielded with less favorable or excellent prognosis in the high RC group (26.8% vs. 46.8%, adjusted OR = 0.31, 95%CI: 0.11–0.85, P = 0.023; or 12.2% vs. 29.0%, adjusted OR = 0.18, 95%CI: 0.04–0.80, P = 0.024, respectively.). Post hoc power analyses indicated that the power was sufficient for favorable outcome (80.38%) and excellent outcome (88.72%) in non-LAA stroke patients. Additionally, RC can enhance the risk prediction value of a poor outcome (mRS scores 3–6) based on traditional risk indicators (including age, initial NIHSS score, operative duration, and neutrophil-to-lymphocyte ratio) for non-LAA stroke patients (AUC = 0.86, 95%CI: 0.79–0.94, P < 0.001). Conclusion: In AIS patients undergoing mechanical thrombectomy, elevated RC was independently related to poor outcome for non-LAA stroke patients, but not to short-term prognosis of LAA stroke patients.
- 5 January 2024
- Published by Springer Nature in Journal of Thrombosis and Thrombolysis
- Vol. 57 (3) , 390-401
Effects of elevated remnant cholesterol on outcomes of acute ischemic stroke patients receiving mechanical thrombectomy
Objective: Large cohort studies provided evidence that elevated remnant cholesterol (RC) was an important risk factor for ischemic stroke. However, the association between high RC and clinical outcomes in acute ischemic stroke (AIS) individuals was still undetermined. Methods: This retrospective study enrolled 165 AIS patients undergoing mechanical thrombectomy in one tertiary stroke center. We divided patients into two groups based on the median of their RC levels (0.49 mmol/L). The modified Rankin Scale (mRS) was used to evaluate the primary outcome 90 days after the onset of symptoms. The mRS scores ≤ 2 and ≤ 1 at 90 days were deemed as favorable and excellent outcomes, respectively. Results: In the overall AIS patients undergoing mechanical thrombectomy, there was no obvious distinction between the high and low RC group at 90-day favorable outcome (41.0% vs. 47.1%, P = 0.431) or excellent outcome (23.1% vs. 31.0%, P = 0.252). In the subgroup analysis stratified by stroke etiology, non-large artery atherosclerosis (non-LAA) stroke patients yielded with less favorable or excellent prognosis in the high RC group (26.8% vs. 46.8%, adjusted OR = 0.31, 95%CI: 0.11–0.85, P = 0.023; or 12.2% vs. 29.0%, adjusted OR = 0.18, 95%CI: 0.04–0.80, P = 0.024, respectively.). Post hoc power analyses indicated that the power was sufficient for favorable outcome (80.38%) and excellent outcome (88.72%) in non-LAA stroke patients. Additionally, RC can enhance the risk prediction value of a poor outcome (mRS scores 3–6) based on traditional risk indicators (including age, initial NIHSS score, operative duration, and neutrophil-to-lymphocyte ratio) for non-LAA stroke patients (AUC = 0.86, 95%CI: 0.79–0.94, P < 0.001). Conclusion: In AIS patients undergoing mechanical thrombectomy, elevated RC was independently related to poor outcome for non-LAA stroke patients, but not to short-term prognosis of LAA stroke patients.