Constant Evolution for the Intelligent AIoT Future: Dahua Technology Unveils Xinghan Large-scale AI Models

NEWS PROVIDED BY

September 19, 2025

Hangzhou, China — Sept. 19, 2025 — Dahua Technology, a world-leading video-centric AIoT solution and service provider, has officially launched its Xinghan Large-scale AI Models — a next-generation, industry-grade AI system that integrates large-scale visual intelligence with multimodal and language capabilities. Developed to address complex real-world environments, Xinghan represents a major leap forward in Dahua’s continuous innovation, empowering intelligent transformation across diverse sectors.

Technological Foundation of Xinghan

With the mission of enabling machines to truly understand the world, the Xinghan model system bridges cutting-edge research with real-world applications. Named after the Chinese word for “galaxy”, Xinghan delivers a full-stack capability matrix powered by edge-cloud synergy, enabling scalable, adaptive intelligence across industries. The upgraded Xinghan architecture comprises three core model series: L, V, and M. The L-series focuses on natural language understanding and interaction; the other two target more specific applications.

V-Series: Xinghan Vision Models

Centered on advanced visual intelligence and video analytics, the V-series streamlines target categories by focusing on key targets (for example: humans, motor vehicles, non-motor vehicles) to reduce model complexity while maintaining high accuracy.

Perimeter Protection: Extended coverage by accurately identifying smaller targets (down to about 20×20 pixels) compared with traditional CNN-based AI models, reducing false alarms and increasing the detection range of large-model cameras.*
WizTracking: A next-generation intelligent tracking algorithm that handles complex occlusions and variations in target posture, delivering substantial improvements in tracking accuracy.*
Crowd Map: Enhanced small-target detection at long distances (up to ~2× farther) and umbrella compensation to improve accuracy in rainy conditions. It offers a multi-fold 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: Leverages situational awareness to analyze spatial and contextual characteristics of a scene, enabling intelligent, automated camera configuration.
AI Rule Assist: Designed for automatic delineation of Perimeter Protection intrusion rules with one-click access, highly accurate scene recognition and automatic analysis.

M-Series: Xinghan Multimodal Models

Multimodal models can simultaneously process and deeply integrate multiple heterogeneous data types (text, images, audio, video). This enhances information processing efficiency, enables more natural human-computer interaction and unlocks a broader spectrum of application scenarios.

WizSeek: Revolutionizes video investigation through natural language search. Describe your target (people, vehicle, animal, item, etc.) and WizSeek retrieves matching footage across recorded archives instantly.
Text-Defined Alarms: Allows users to define alarms by describing them in natural language, lowering development thresholds and enabling fast, flexible, scalable configuration for diverse, real-world scenarios.

For more information about the Xinghan Large-scale AI Models, please visit https://www.dahuasecurity.com/Products/keyTechnologies/modalXH.

*Results are based on standard setup and testing environment.