TECH OFFERS

This technology offer is a low cost, smart thermal sensor with proprietary thermal imaging technology. The thermal sensor has edge AI capabilities comprising on-board computer vision algorithms, enabling advanced applications such as human tracking, hotspot tracking and fall detection. The technology has native low resolution so it is privacy non-intrusive and works in all lighting conditions, making it the technology to go for when it comes to monitoring of people, animals, hazardous hotspots and assets in all those spaces where privacy and/or cost are a concern.

Most existing training applications offer good programmes for guiding users to achieve individual fitness goals, some even come with guided video workouts led by professional trainers. However, such applications lack or have limited capability to assess whether the correct posture is maintained during exercise - poor posture can reduce exercise effectiveness and may even cause injury, e.g. arched back during push-ups. This solution is a synergistic combination of video/image processing, human pose recognition, and machine learning technologies to deliver a solution that addresses the twin challenges of accurate count and correct execution of exercises in an automated manner, without having to wear any additional hardware/sensors. The software-only solution is able to advise users on the correct execution of repetitive movement sequences, e.g. sit-ups, and push-ups, and is deployable on a wide range of affordable camera-enabled hardware devices such as mobile phones, tablets, and laptops, and it can be easily integrated into existing applications to enhance functionality. It is applicable to the sports and healthcare industry to help users perform exercises correctly and effectively in an unencumbered manner.

Existing publicly available datasets, such as COCO, are built from the ground up to be general-purpose and therefore lack domain specificity. When such public datasets are used to train deep learning models for industrial use-cases and applications, e.g. detection of electronic components, they often result in sub-par performance caused by the disparity between objects typically found in industrial environments and data residing in public datasets. This disparity requires significant effort in pixel-level supervision (annotation), where each pixel, per frame, has to be annotated manually to make up for the difference in training data to improve model performance This solution is a deep-learning-based technique for instance segmentation in industrial environments intended to reduce the effort cost of annotation from pixel-level to video-level. With instance segmentation, the goal is not just to detect and localise objects within a scene, but also to determine the different classes and number of instances (or recognising more of the same type objects as different). This aids scene understanding and the resulting model can be deployed for productivity measurement or process improvement. Incremental learning is used to ensure that only the parts of the model that need to be updated with new data are changed, thus reducing the amount of time taken for re-training and model updates.

With the rapid pace of digitalisation, many existing systems and processes are becoming increasingly complex. Many users find themselves struggling to achieve their desired outcomes due to a wide variance in digital proficiencies; what is intuitive to one user may not be intuitive to another. Simply put, it is not possible to build a User Interface (UI) that is completely intuitive for every user profile i.e. no one-size-fits-all interface. Similarly, many Frequently-Asked-Questions (FAQs) and user guides are poorly maintained or are written in a manner that is too generic with little to no consideration of a user's role or level of proficiency. This technology offer is a no-code solution that can be deployed on websites to provide just-in-time (JIT) tutorial-style overlays which bridge the gap between digital workflows and human usage. These customisable overlays serve as guided walkthroughs to simplify employee onboarding and/or provide external users with a curated customer experience (CX). With this solution, highly re-useable, step-by-step Standard Operating Procedures (SOP), user manuals, and interactive guides can be easily created to provide clear, simple instructions for end-users to receive assistance when needed, this in turn, improves their understanding of proper software product usage and provide a painless user interface experience.

While interest and demand for blockchain-related technologies continue to gain popularity and spurs the exponential growth in adoption of blockchain in areas such as payment, documents and digital identities, not just in finance, but in industries such as logistics, supply chain as well. Many emerging areas that rely on blockchain as a core technology lack the manpower needed to sustain budding development, this lack of technical skillset required for blockchain development is the primary hurdle to successful blockchain application development - less than 1% of the tech workforce is skilled or competent in blockchain-related development. The characteristic of blockchain technology, which enables a permanent record of digital information such that it cannot be modified by any single entity renders it well suited as a digital ledger of online transactions. As such, blockchain is a core technology for many emerging areas such as Decentralised Finance (DeFi) and Decentralised Autonomous Organisation (DAO). This technology offer consists of a platform tool and a set of zero-configuration REST APIs that abstract away the complexity of blockchain technology and enables any developer to easily build blockchain-based applications or integrate blockchain functionality into their existing systems. Intended as a low-code platform, it addresses the skills gaps traditionally required for blockchain development and deployment and allows companies to realise their blockchain ideas, enhance business operations and expand solution offerings.  

The current clean meat technologies grow lab meat with conventional 2D cell culture. However, the conventional cell culture technique has an overall low yield of cells, as the cells are restricted to growth on surface areas.  A new 3D scaffolding method has been developed to overcome this problem with the use of microcarrier beads that provide cells with additional surface area to attach onto and proliferate. The microcarrier beads are suspended in the cell culture thus maximizing the 3D volume of the cell culture, leading to an increased yield. A microcarrier type has been identified to yield the highest number of porcine cells. The conditions of the cell culturing process have been optimised to improve the cell viability in a 3D environment Companies interested in cell-cultured meat development could consider using this method to grow cell-cultured meat at a larger scale with a potentially lower cost of production. The technology developer is seeking companies that are keen to scale up lab-grown meat applications. 

Many existing smart city solutions only show the impact of urban development, but few show the impact that urbanisation imposes on daily activities and long-term outcomes such as population obesity and job availability/accessibility. In short, such solutions show the activities e.g. large crowds are visiting the neighbourhood park, that are happening in real-time (what), the location (where), and the time that they occur (when), but do not have the ability to include data that makes it possible to explain the reason for such activities (why). In order to bring about any intervention or identify missed opportunities, understanding the reason behind such activity is vital. This technology utilises data on city infrastructure systems to help users understand how and where the built environment creates a set of physical constraints that influence what planned and unplanned activities are possible, and in turn how this influences long term outcomes including health and climate change. This technology imports, translates and combines datasets into spatialised models which are used to generate analytics outputs. These outputs include a comprehensive explanation of the way streets, pedestrian networks, public transport and land use interact with each other. In this manner, socio-economic and/or demographic datasets can be linked, enabling people and places to be combined in a single analytical model.

This technology offer presents an autonomous materials handling system. The technology could reduce manual labor requirements and increase working efficiency. The solution consists of a Lidar Elevator Stand (LES) system, which will trigger the autonomous actuators (e.g., trolley puller, tray return robots) via the robot command center whenever no goods (e.g., trolley, tray, and stock) is detected in the designated area. Currently, the technology has been demonstrated in autonomous trolley return solutions. Generally, trolley replenishment requires deploying manual labour to monitor the available quantity at the trolley bay and replenishing it by physically operating an electric trolley puller to transport the new stack of trolleys. Therefore, the system was developed to solve the problem by triggering an autonomous trolley puller to replenish the new stack of trolleys whenever the trolley quantity is depleting. The system can be further customized and repositioned based on clients’ requirements.

The proliferation of e-commerce, ride-hailing and food-delivery services have fueled the need for more accurate and reliable estimation of delivery times. The current common estimation of delivery time is based on Estimated Time of Arrival (ETA) which relies on route distance that is calculated between the origin and the desired destination. It only considers the duration from pick up to drop off, and does not consider the additional time needed for preparing and offloading the goods. This technology offer is a Machine Learning (ML) model that is able to calculate the stop duration (job completion duration), which together with the ETA, provides the Estimated Time of Completion (ETC). This ML model is for Singapore use only.