Research Paper

Why It Matters

Accurate and reliable visual tracking is critical for applications like drone navigation and security surveillance, where losing track of targets due to occlusions or similar distractors causes failures. OA-VAT improves tracking stability and precision, reducing operational risks and enabling continuous monitoring in dynamic, cluttered environments. This enhances automation and safety across industries relying on visual tracking.

Potential Customers & Pain Points

• Drone manufacturers – Need reliable target tracking despite occlusions
• Security companies – Require stable surveillance in crowded scenes
• Robotics firms – Demand robust visual navigation under dynamic conditions
• Autonomous vehicle developers – Face challenges with occlusion and distractor confusion.

Research Paper

Why It Matters

Adversarial patches pose a critical security risk by causing misclassifications in vision AI systems, impacting applications from autonomous vehicles to surveillance. STRAP-ViT offers a practical, deployable solution that preserves model accuracy while protecting against these attacks, enabling safer and more reliable AI deployments at scale.

Potential Customers & Pain Points

• Autonomous vehicle manufacturers – Vulnerable to adversarial attacks causing misperception
• Security and surveillance firms – Need robust image recognition against physical tampering
• AI model providers – Require plug-and-play defenses without retraining costs
• Cloud AI service providers – Demand scalable low-overhead adversarial protection.

Research Paper

Why It Matters

Fingerprint recognition systems rely on accurate minutiae extraction, which traditionally requires multiple processing steps and specialized tools. LEADER simplifies and accelerates this process with a single efficient model, improving accuracy especially on latent fingerprints and reducing computational costs. This enables faster, more reliable biometric authentication and forensic analysis at scale.

Potential Customers & Pain Points

• Biometric security providers – Need accurate and fast fingerprint analysis
• Law enforcement agencies – Require robust latent fingerprint extraction
• Mobile device manufacturers – Demand efficient on-device fingerprint recognition
• Forensic labs – Seek reliable cross-domain fingerprint matching.

Research Paper

Why It Matters

Suspiciousness estimation is vital for proactive threat detection in crowded and complex public spaces. This solution improves accuracy and interpretability of risk assessments, enabling faster and more reliable security responses. It scales across diverse environments, supporting safer public venues and critical infrastructure.

Potential Customers & Pain Points

• Public safety agencies – Need accurate real-time threat detection
• Airport and transit security – Require scalable surveillance analytics
• Private security firms – Demand interpretable suspiciousness alerts
• Smart city operators – Seek integrated risk assessment tools.

Research Paper

Core Innovation

This paper introduces Vi-SAFE, which integrates an enhanced YOLOv8 with a Temporal Segment Network to detect violence in videos efficiently. It uses a lightweight GhostNetV3 backbone and EMA attention to reduce computational cost without sacrificing accuracy. The separate training of detection and classification modules improves performance on small-scale and complex surveillance targets.

Potential Customers & Pain Points

• Public Safety Agencies Needing Real-Time Violence Alerts
• Surveillance System Providers Seeking Efficient AI Integration
• Smart City Developers Requiring Scalable Violence Detection
• Security Firms Facing High Computational Costs
• Event Organizers Wanting Automated Threat Detection

Research Paper

Core Innovation

This paper introduces Multi-Turn Text Generation (MTG) to create diverse pseudo-labels without manual captions and Multi-Turn Text Interaction (MTI) to refine user queries dynamically. These modules together enable annotation-free, dialogue-based person retrieval that handles ambiguous or incomplete descriptions better than prior methods.

Potential Customers & Pain Points

• Surveillance agencies needing scalable person search
• Security firms requiring accurate image retrieval
• Law enforcement handling vague or incomplete descriptions

Research Paper

Core Innovation

This paper presents TACR-YOLO, which integrates Coordinate Attention and Task-Aware Attention modules to enhance feature representation and resolve task conflicts. It also introduces a Strengthen Neck Network to improve multi-scale fusion and optimizes anchor boxes and bounding box regression for better detection accuracy. These innovations collectively improve detection of small and abnormal human behaviors in real time compared to prior models.

Potential Customers & Pain Points

• Security Companies Needing Accurate Behavior Detection
• Surveillance System Providers Facing Small Object Detection Challenges
• Public Safety Agencies Requiring Real-Time Alerts