With significant increases in chip computing power and the rapid evolution of algorithms, AI is widely applied in various areas such as autonomous driving, smart transportation, smart manufacturing, smart cities, and smart healthcare. For smart manufacturing, AI not only benefits manufacturing process optimization, productivity, and product quality detection, but also enables applications of people access control, production environment monitoring, and industrial worker safety.
Monitoring of public places is also critical for smart cities. It not only facilitates daily road inspections using mobile vehicles for governance but also effectively reduces criminal incidents, enhancing residents' living security.
There are numerous existing smart surveillance solutions in the market, but in Linker Vision's opinion, those legacy solutions have two major drawbacks difficult to get rid of: First, training the initial AI model costs a huge amount of manpower and time to collect a large amount of data and do labeling work, which would be an obstacle to deploying AI applications, leading to a delayed schedule.
Second, unable to keep optimizing AI models to promptly handle variations in the ambient environment. It's difficult to improve the generalization of overall recognition performance, and successful AI cases cannot be easily replicated in increasingly complex environments.
"We've accumulated a lot of experience in the area of autonomous driving and provided high-complexity data processing services to international OEMs such as Sony and Bosch." Dr. Paul Shieh, the Founder & CEO of Linker Vision said. "For smart manufacturing, Linker Vision has offered end-to-end intelligent monitoring solutions to big names such as TSMC and Formosa Plastics."
Recognizing the strong demand for AI solutions in applications of smart manufacturing and smart cities, Linker Vision introduces a continuous learning AI platform, DataVerse, with its exclusive Large Vision Model as the kernel. By partnering with edge computing hardware providers and leveraging Amazon SageMaker to speed the process of AI model development and iteration, improving efficiency by more than 10x. In the meantime, combined with AWS Panorama, it can effectively save the time of deploying AI models at the edge by more than 50%.
Linker Vision built the standardized, general-purpose VisionAI platform, and perfectly integrated the software and hardware of 5G AIoT. It's for the global market, and supports continuous learning and iterative optimization, to help enterprises significantly reduce the cost and time for deploying intelligent surveillance solutions.
Linker Vision's 5G AIoT solution can be applied to strengthen surveillance in factories, including equipment monitoring against issues of pipeline corrosion, and gas leaks. It also supports the management of restricted areas, early detection of fire and smoke, identification of abnormal employee behavior, and monitoring of other potential hazards and security risks.
For the usage of smart manufacturing, their existing customer spans industries such as semiconductors, petrochemicals, steel, manufacturing, logistics, and environmental companies. Its application scope also extends to smart cities, keeping close collaborations with the municipal governments of Kaohsiung and Tainan.
Speed the Deployment of Smart Surveillance
Linker Vision concentrates on the development and application of Large Vision Model, its 5G AIoT solution has three key features: Acceleration, Generalization, and Service Replicability. In terms of Acceleration, the company provides an efficient Auto-Labeling tool and large-scale image models, which can help businesses shorten development cycles for fast building and deploying visual AI applications.
To Generalization, the approach involves utilizing multimodal AI with continuous learning capabilities, enabling VisionAI models to be widely applied in different environments to meet the diverse usage requirements of various scenarios. Service Replicability emphasizes making solutions and services replicable for applications across multiple domains.
Melody (Ruei-An) Liu, Global Business Development Manager at Linker Vision, explained that the 5G AIoT solution is a cloud-edge integration solution encompassing advanced image analysis, data management, auto-labeling, model monitoring, and more. It provides fast deployment and surveillance capabilities, allowing businesses to be informed of all the conditions of their workplaces.
And because Linker Vision train and optimize the Vision AI models on the Amazon Web Services (AWS) platform. Users can easily get the latest version of VisionAI on the AWS Marketplace for future requirements of different use cases.
"It's our pleasure to be invited to participate in the Startup Terrace Kaohsiung AWS Joint Innovation Center (JIC). By leveraging AWS' cloud technologies to accelerate product development, we aim to provide customers with more diverse services, and see it as a strong support for our expansion to the global market." Liu added.
Based on all the rich and successful cases in the Taiwan market, Linker Vision is set to vigorously expand into the global market in 2024. By working with local partners, the company plans to introduce its 5G AIoT solution to Southeast Asian countries such as Vietnam, Thailand, and Singapore, as well as the European and North American markets.
As a part of the strategy, the company will announce a whole new product at CES 2024, and go on to promote it at MWC 2024. By demonstrating the features and advantages of its solutions, they hope to attract more interest and opportunities for collaboration.
Large Vision Model speeds up the process of AI model development and iteration, improving efficiency by more than 10x
Photo: Linker Vision