Technological Foundation of Xinghan
With a mission to enable machines to truly understand the world, the Xinghan model system continues to evolve by bridging cutting-edge research with real-world application. Since initiating visual attention research in 2017 and redefining AI in the security domain with the launch of its visual cognition foundation model in 2023, Dahua has demonstrated ongoing technological leadership in AIoT innovation.
Named after the Chinese word for “galaxy”, the newly upgraded Xinghan system delivers a full-stack capability matrix powered by edge-cloud synergy, enabling scalable, adaptive intelligence across sectors. Driven by visual analytics and industry-specific demands, Xinghan integrates multimodal intelligence with deep domain expertise. This development has resulted in three core series: the L, V, and M and models. The L model focuses on natural language understanding and interaction, while the other two models tackles more specific applications:
V-Series: Xinghan Vision Models
Centered on advanced visual intelligence and video analytics, this series streamlines target categories by focusing on key targets (e.g. humans, motor vehicles, and non-motor vehicles) to reduce model complexity while preserving high accuracy. Its key features include:
Perimeter Protection
- Perimeter Protection: The coverage and range of perimeter detection are extended by accurately identifying smaller targets (even down to 20×20 pixels) compared to traditional CNN-based AI models, reducing false alarms and increasing the detection range of large-model cameras.*
WizTracking
- WizTracking: It offers a next-generation intelligent tracking algorithm that can handle complex occlusions and variations in target posture, achieving 50% improvement in accuracy.*
Crowd Map
- Crowd Map: It significantly enhances small-target detection at long distances (up to 2× farther) and features umbrella compensation, improving accuracy by 80% during rainy conditions*. It also offers a 2.5× increase in analysis range, supports detection of up to 5,000 people, and provides robust performance in dense crowds and low-light environments.*
Scene Adaptive – AI WDR
- Scene Adaptive – AI WDR: It leverages situational awareness to analyze both spatial and contextual characteristics of a scene, enabling intelligent and automated camera configuration.
AI Rule Assist
- AI Rule Assist: It is designed for the automatic delineation of Perimeter Protection intrusion rules, offering one-click access, highly accurate scene recognition, automatic analysis, and more.
M-Series: Xinghan Multimodal Models
Multimodal models are advanced AI systems capable of simultaneously processing and deeply integrating multiple heterogeneous data types (e.g. text, images, audio, and video). Their core capability lies in leveraging advanced cross-modal representation alignment and joint semantic comprehension techniques to bridge gaps between modalities, achieving deep semantic correlation and collaborative understanding across diverse data sources. This capability not only significantly enhances the efficiency and richness of information processing but also enables more natural human-computer interaction and unlocks a broader spectrum of application scenarios. Its notable features include:
WizSeek
- WizSeek: This feature revolutionizes video investigation through natural language search. Simply describe your target (e.g. people, vehicle, animal or item, etc.) and WizSeek instantly retrieves matching footage across recorded video archives. It offers a wide-range, instant, user-friendly and accurate target searching method.
Text-Defined Alarms
- Text-Defined Alarms: It enables the creation of custom alarm rules through simple text input – no coding, manual development, or costly, time-consuming processes required. It allows users to define alarms by simply describing them in natural language, which significantly lowers the development threshold, and enables fast, flexible, and scalable configuration tailored to diverse real-world scenarios.
Looking Ahead: Building the Future of Intelligent AIoT
Moving forward, Dahua will continue to enhance the Xinghan Large-scale AI Models to meet the growing demand for intelligent transformation. By working closely with ecosystem partners and customers worldwide, Dahua aims to expand the application of large models in real-world scenarios, fostering new momentum in digital public security, smart transportation, energy management, and enterprise-level innovation.
For more information about the Xinghan Large-Scale Models, please coordinate with your local Dahua representative or visit the official webpage here.
*Results are based on standard setup and testing environment
