BEIJING, July 27, 2026 /PRNewswire/ — WiMi Hologram Cloud Inc. (NASDAQ: WIMI) (“WIMI” or the “Company”), a globally leading technology provider, has delved into the interdisciplinary field of quantum computing and computer vision to carry out research on quantum visual tracking algorithm technology. Deeply integrating the core characteristics of quantum computing with classic visual tracking frameworks, this algorithm reconstructs the full-process logic of algorithm training and detection, achieving technological innovations at the underlying computing power and algorithm architecture levels. Boasting both theoretical novelty and engineering implementation value, it effectively resolves the trade-off between accuracy and efficiency that plagues traditional algorithms in complex scenarios, delivering an all-new technical solution for high-performance visual tracking systems.
The quantum visual tracking algorithm developed by WIMI abandons the linear computational logic of classic visual tracking. Based on the operating mechanisms of quantum computing, it constructs a closed-loop architecture consisting of two phases: quantumized training and quantum parallel detection. Core technologies including quantum state encoding, quantum parallel computation and quantum superposition solving are embedded throughout the entire workflow, transforming the feature extraction, model training and target detection paradigms adopted by conventional algorithms. Compared with classic algorithms, this architecture fully unlocks the core strengths of quantum computing in high-dimensional spatial computation and parallel batch processing, bringing comprehensive improvements in model training efficiency, target detection accuracy and adaptability to dynamic scenarios.
The core innovation in the algorithm training phase lies in the quantum-state encoded ridge regression classifier. To address the drawbacks of classic ridge regression models — low efficiency of parameter optimization and severe multicollinearity interference when handling high-dimensional visual features and large-scale datasets — a quantumized upgrade has been implemented. Classic ridge regression introduces regularization terms to constrain model parameters and realize stable fitting of multicollinear data. Nevertheless, training with massive video feature datasets entails an extensive parameter search space, lengthy iterative optimization cycles and extremely high computing power consumption. The proposed algorithm maps and encodes the optimal fitting parameters of the ridge regression model onto the amplitude dimension of quantum states. Leveraging the natural high-dimensional spatial representation capability of quantum states, it enables high-dimensional structured storage and extraction of visual features. Meanwhile, by harnessing quantum parallel computing properties, it breaks the single linear iterative parameter optimization mode of classic algorithms, synchronously completing feature screening, parameter iteration and model fitting within the quantum high-dimensional space. This innovation drastically shortens the training cycle for large-scale visual datasets and markedly enhances the stability and precision of model parameter estimation, with more prominent computing power gains especially in training scenarios involving massive video samples and high-dimensional image features.
In the video frame target detection phase, the algorithm invokes the trained quantum-state ridge regression classifier to perform batch classification and detection on all candidate target image patches within a single video frame. Unlike the serial detection mode of classic algorithms that traverses and matches targets one by one, this quantum algorithm leverages the core property of quantum superposition to accommodate all detection tasks for candidate patches simultaneously, completing feature matching and result solving for all target samples in parallel. During detection, the classifier generates new quantum states and encodes the target response results of all candidate patches via quantum amplitudes, delivering a parallel detection effect of one computation with full-domain outputs. Additionally, the algorithm exhibits excellent numerical adaptability. When the image data matrix has a low condition number and high data complexity, it delivers logarithmic scaling performance, achieving exponential computational acceleration compared with state-of-the-art classic visual tracking algorithms. It fundamentally addresses the issues of detection latency and excessive computing load in high-complexity scenarios.
Against traditional visual tracking algorithms, the quantum visual tracking algorithm researched by WIMI achieves overwhelming technical advantages across four dimensions: computing power efficiency, tracking accuracy, scenario adaptability and resource utilization, meeting the demands of various high-end real-time visual tracking scenarios. In terms of computing power and real-time performance, the algorithm breaks through the upper limit of classic computing power by virtue of quantum superposition and parallel computation, enabling ultra-fast processing of large-scale video data and high-frequency dynamic frame sequences to satisfy the low-latency requirements of real-time tracking applications. For accuracy and robustness, the high-dimensional spatial feature representation capability of quantum systems can finely distinguish target features from complex background interferences, effectively mitigating tracking drift and target loss induced by light and shadow variations, cluttered backgrounds and target occlusion, and greatly boosting tracking stability. Furthermore, the quantum algorithm restructures computational logic to streamline redundant calculation steps. Relying on the efficient data encoding characteristics of quantum states, it significantly cuts down storage overhead and operational resource consumption for visual data. Its lightweight computational architecture is perfectly compatible with edge computing, embedded terminals and other devices with limited computing power. In terms of scenario adaptability, the algorithm can self-adapt to diverse extreme environments such as dynamic backgrounds, high-speed moving targets and complex light and shadow conditions, demonstrating strong universal compatibility. The quantum visual tracking algorithm effectively makes up for the application shortcomings of traditional visual technologies and can be deployed to empower core scenarios including high-end human-computer interaction and intelligent security monitoring, possessing extremely high practical application value.
The quantum visual tracking algorithm developed by WIMI realizes in-depth integration between quantum computing technology and the field of computer visual tracking, breaking through the technical limitations of classic visual algorithms and blazing a new trail for the iterative upgrading of computer vision technologies. With continuous iteration of quantum technologies and constant optimization of quantum algorithm systems, quantum visual technologies will further unleash their computing power advantages. The quantum visual tracking algorithm can be deployed in industrial scenarios such as intelligent robot navigation, autonomous driving visual perception and high-precision industrial visual inspection. In the future, quantum-empowered computer vision technologies will drive sustained industry development toward higher precision and higher-efficiency intelligence.
About WiMi Hologram Cloud Inc.
WiMi Hologram Cloud Inc. (NASDAQ: WIMI) focuses on holographic cloud services, primarily concentrating on professional fields such as in-vehicle AR holographic HUD, 3D holographic pulse LiDAR, head-mounted light field holographic devices, holographic semiconductors, holographic cloud software, holographic car navigation, metaverse holographic AR/VR devices, and metaverse holographic cloud software. It covers multiple aspects of holographic AR technologies, including in-vehicle holographic AR technology, 3D holographic pulse LiDAR technology, holographic vision semiconductor technology, holographic software development, holographic AR virtual advertising technology, holographic AR virtual entertainment technology, holographic ARSDK payment, interactive holographic virtual communication, metaverse holographic AR technology, and metaverse virtual cloud services. WiMi is a comprehensive holographic cloud technology solution provider. For more information, please visit http://ir.wimiar.com.
Translation Disclaimer
The original version of this announcement is the officially authorized and only legally binding version. If there are any inconsistencies or differences in meaning between the Chinese translation and the original version, the original version shall prevail. WiMi Hologram Cloud Inc. and related institutions and individuals make no guarantees regarding the translated version and assume no responsibility for any direct or indirect losses caused by translation inaccuracies.
SOURCE WiMi Hologram Cloud Inc.