At the core of our services are proprietary image recognition engines, which are powered by our retail domain knowledge and deep learning. Our services establish a full-stack solution that can meet and surpass the needs of our FMCG supplier and retailer clients, all the way from image and data collection to reporting, visualizing, aggregating rich sets of retail execution KPIs, and enabling advanced analytics on this Big Data.
Our engines are specifically tailored for retail scenes, for packaged products and their taxonomy, whereas off-the-shelf image recognition frameworks and APIs lack any domain specific adaptation. This makes our Image Recognition (IR) technology superior to its counterparts in terms of accuracy, repeatability and ability of handling diverse use cases.
Vispera Recognition Engines are reinforced by the most sophisticated panorama reconstruction and image segmentation algorithms for retail scene parsing, which makes our image rejection rates record-low, well below 1% over all submitted images.
Vispera Recognition Engines have high tolerance to adverse picture-taking conditions such as non-frontal views of shelves taken in narrow aisles or pictures with severe glass reflections from coolers with doors closed.
Vispera Recognition Engines support SKU level and brand level product recognition on a sheer amount of categories and package types. Thanks to their learning capabilities under small sample sizes, our engines can be swiftly trained and made ready for deployment. For training our engines, we do not require high resolution shots taken at ideal studio environment, just a single photo for each SKU and a handful of retail scene images depicting the target products are sufficient for deploying our visual models.
The baseline performance of Vispera Recognition Engines starts from 93% accuracy for SKU level product recognition, and exceeds above 96% depending on the product category as more training data accumulates for continued machine learning. Engines are further powered by our proprietary confidence-enabled human-in-the-loop system, which enables us to efficiently deliver above 99% SKU level recognition accuracy. This makes us the only IR solution provider in the retail domain, who can commit to near-perfect performance levels.
VISPERA RECOGNITION ENGINE PERFORMANCE
