Lab

viAct - AI Video Analytics for Safety & Construction Management Software

Institution: Independent Researcher

About the lab

viAct is a leading ESG-focused AI company that provides “Scenario-based Vision Intelligence’ solutions for the AEC industry across Asia, Europe & Middle East. With hundreds of deployments since its inception in 2016, viAct’s proprietary AI algorithms have evolved successfully providing extremely granular insights in jobsites by transforming vision to practical actions. Our 30+ pre-built AI modules enable the stakeholders to reduce the number of accidents, optimize costs and track environmental non-compliances. As a 2020 Top 50 Global ConTech Startup, viAct’s disruptive AI navigation solution is one of a kind approach to manage man-made environments in a far smarter way than humans do!

Featured research (12)

Most of the AI-based applications are built on the foundation of deep learning methods. In the ConTech ecosystem, large scale user data are collected through vision intelligence which is proportionally used for training the AI for various scenarios. In this context, it is observed that for building any AI module, a massive data collection is the prime necessity for deep learning which accompanies inevitable privacy issues. Highly sensitive user data such as photos and videos are indefinitely with the companies which collect them and user cannot delete it or restrict its usability. Thus, vision intelligence powered AI which is popularly used in the ConTech ecosystem are potential subject to legal and privacy matters. GDPR regulations are stringent in this sector because with the inclusion of AI in the construction sector there is a rise in risks of privacy damages. However many startups and large companies have set good example of accuracy maintaining privacy norms together. viAct (Hong Kong) is one of such world class scenario based ConTech startups known for its privacy ensuring platform. viAct has taken steps like blurring and masking of human faces, encryption of stored data, privacy preserving deep learning for computer vision and edge AI for computing in order to mitigate such privacy issues. The presented case study of viAct’s AI thus showcases a good example of responsible AI.
Construction is an activity that fulfills one of the basic needs of humans (i.e., shelter). However, this industry is also known for its excessive waste generation, which impacts the environment if not disposed of appropriately. Much of the waste consists of harmful material; when dumped in the landfills, this leads to gradual leaching of various undesirable ions into the groundwater, causing water quality deterioration. Such leachate-rich water when used by humans for various purposes causes diseases and deformities. Thus, to improve the ecological civilization and to promote the overall ESG proposition in the construction industry, artificial intelligence (AI) is a suitable solution. This chapter puts forward a case study of an AI-based ConTech startup, called viAct that developed and tested AI modules for monitoring waste generation and disposal at construction sites. The AI modules are trained and tested for their efficacy in various construction sites for illegal dumping detection and classification of different types of waste material before they are discharged in landfill areas.
This paper is an attempt at explaining what carbon credits are and how AI helps in carbon emission monitoring and carbon credit management in construction taking the case study of viAct (A ConTech Startup from Hong Kong). viAct's scenario-based AI has been designed to play a significant role in measuring, monitoring, tracking, predicting and reducing carbon emissions. Further, its solutions like fleet management and its AI modules such as Air Quality Detection, C&D Waste Classification, and Illegal Dumping Detection has helped the construction companies to optimize construction machinery usage; monitor air quality and C&D wastes and detect illegal dumping of these wastes, respectively. In addition to this, the auto-documentation and analytics capabilities of viAct's AI monitoring platform-viHUB is an exclusive solution helping stakeholders in managing their carbon credits and well as empowering their carbon credit trading in a holistic manner.
viAct's 360° progress tracking camera is a special AI enabled device build for better accuracy of progress tracking in construction jobsites. The device is a small handy camera that any site inspector can carry during his walk around the jobsite. The camera is powered with unique wall texture classification algorithms preceded by wall segmentation algorithm that makes it different from other technologies used in construction jobsite for progress monitoring. The research includes an automated pipeline built up for experimentation of different feature extraction algorithms, supervised and unsupervised learning models, and customized data to select the most compatible combination for wall texture classification. The wall texture classification is set to around 85% on every prediction. The pipeline includes data preparation, feature selection, model selection, hyper-parameters tuning, validation and prediction. The unique AI of the camera makes the process of progress tracking faster, accurate and reliable when compared to other traditional as well as digitized jobsite tracking systems. Thus, viAct's newly launched 360° progress tracking camera is tailor-made for construction sites to prevent tons of manual documentations and to enable real-time alerts with lesser efforts for better transparency in the workflow.
The paper presents research insights on amalgamation AIoT and Robotic technology for construction site monitoring. For this purpose the paper studies the special case of viAct's smart cloud controlled robotic system with embedded camera that enables real time capturing of images, detecting various safety non-compliances by workers for maintaining a safe construction ecosystem. Furthermore, the automatic wheel base carries LiDar that can detect the objects in the form of point clouds generating a as-is model which is compared with the as-planned BIM model. This helps in regular day to day progress tracking of the real time scenario helping contractors and owners to have human prejudice free inspection of their construction sites. viAct's smart robotic system working in a collaborative robot background, is thus a value addition to the viAct's existing monitoring solution based on AIoT, making the entire monitoring more accurate, optimal and error free.

Lab head

Gary Ng
Department
  • Research
About Gary Ng
  • R&D in Computer Vision in Construction, AI application in Construction Safety, AI application in Construction Productivity, AI Automation for Autonomous Monitoring in Robotics