Fizyr designs, builds, installs, and maintains the world’s best vision software product for automated picking and placing in harsh logistics environments. Our deep learning algorithms are used to enable automated handling of unknown objects varying in shape, size, color, material, and stacking, being picked from (conveyed) bulk. Fizyr’s product is applied in production all over the world by leading system integrators and end-users, in online retail, warehousing and parcel services.
We started in 2014 as ‘Delft Robotics’, a TU Delft spin-off founded by Professor Martijn Wisse, as a system integrator building robotic picking cells. In 2016, we participated in and won both competitions of the Amazon Picking challenge, mainly due to our track record in applying deep learning. This led to the company pivoting towards a deep learning computer vision software-only product and re-branding to Fizyr in 2017. Our Plug-and-Play modular product allows for a smooth integration with any system. We give integrators the freedom to choose the best hardware for their picking cell and we help them configure it, while facilitating cost, robustness, and speed optimization. Our algorithms provide over 100 grasp poses each second, including classification to handle objects differently, but also perform quality controls and detect defects to prevent damaged items from being placed on a sorter.
Leading integrators across the globe acknowledged our product’s high quality and used it in production. This is what makes us truly unique; we were trusted as a strategic partner and grew successfully bootstrapped since the beginning. Up to this day, we remain dedicated to creating true value for our partners, so that they can create cutting-edge logistics solutions for their clients.