TaQadam offers a solution to build an alternative data set on assets with computer vision and human validation
How it Works
We work on building data sets using classification and object detection Machine Learning models to give a better understanding of footprint, object count (structure class, sizes, geo-location). The data is labeled in iterative cycles: training, testing, validation sets to arrive to a full solution on Asset & Commodities.
For better asset condition and asset performance monitoring we bring a hybrid model: use of computer vision along side human validation and meta tags to bring