A5080239367- Artificial Immune Systems Applications
- Anomaly Detection Techniques and Applications
- Network Security and Intrusion Detection
- CCD and CMOS Imaging Sensors
- Ferroelectric and Negative Capacitance Devices
- Hand Gesture Recognition Systems
- Human Pose and Action Recognition
- Advanced Memory and Neural Computing
- Diabetic Foot Ulcer Assessment and Management
University of Hong Kong
2022-2024
Weakly supervised detection of anomalies in surveillance videos is a challenging task. Going beyond existing works that have deficient capabilities to localize long videos, we propose novel glance and focus network effectively integrate spatial-temporal information for accurate anomaly detection. In addition, empirically found approaches use feature magnitudes represent the degree typically ignore effects scene variations, hence result sub-optimal performance due inconsistency across scenes....
Event-based vision represents a paradigm shift in how information is captured and processed. By only responding to dynamic intensity changes the scene, event-based sensing produces far less data than conventional frame-based cameras, promising springboard new generation of high-speed, low-power machines for edge intelligence. However, processing such dynamically sparse input originated from event cameras efficiently real time, particularly with complex deep neural networks (DNN), remains...
To reduce the injury and fatality of special group pedestrians in rapid traffic flow, engineers can utilize artificial intelligence to detect human bodies postures. The design enables identification existence intersection identifies their postures with NVIDIA CUDA Deep Neural Network (CuDNN). This mechanism helps authority devise an advanced algorithm manipulate lights allow these accommodation-needed people cross road time when they are intersections. By experiments, Zaixin has found that...
Weakly supervised detection of anomalies in surveillance videos is a challenging task. Going beyond existing works that have deficient capabilities to localize long videos, we propose novel glance and focus network effectively integrate spatial-temporal information for accurate anomaly detection. In addition, empirically found approaches use feature magnitudes represent the degree typically ignore effects scene variations, hence result sub-optimal performance due inconsistency across scenes....
Human pose estimation (HPE) has been gradually applied to our daily life. It's significant design a simple yet effective model structure for real-time HPE. Several backbones are available estimation, but many of them imprecise, complex, and mislocalized when it comes reconstruction. In order narrow the gaps, several recent studies have demonstrated that deconvolution reconstruction is highly in achieving high levels accuracy. Using current popular backbones, we re-analyze reconstruct models....