Blippar is the market-leading image recognition and augmented reality platform with the power to bring the physical world to life via the camera of a smart device.
We are now seeking a computer vision engineer to join our R&D team that brings our technology to the next level, and that builds some of the core technologies that integrate in the app and our backends, using modern C++ to achieve the best performance.
Blippar is continuously improving its Augmented Reality and image recognition technologies. Robustness and speed of our app are two key criteria that make it easier for people to experience AR. It is of crucial importance that our customers can enjoy a consistent and seamless experience, regardless of what object they point their camera at. You will be researching as well as implementing the latest technologies in the fields of image retrieval, classification and detection, natural feature tracking, SLAM, and feature-based 3D reconstruction. You will also be responsible for building and delivering new features on our roadmap.
The ideal candidate will be flexible, self-motivated, and loves performing product-oriented research as well as writing excellent code. You will be reporting to the Head of Computer Vision R&D, joining a small team of enthusiastic developers that are stretching mobile and server-side technologies to the limits of the available device capabilities.
Being part of Blippar’s close-knit, passionate, forward-thinking team, you will have the opportunity to shape a brand new medium and the way in which people all over the world use it.
As computer vision R&D engineer, we expect you to have PhD level academic experience in computer vision with a strong focus on machine learning, as well as a proven track record of practical experience in areas such as natural feature tracking, multi-view geometry, 3D reconstruction, image retrieval, image classification and object detection. Hands-on experience with deep learning for large-scale recognition, including familiarity with frameworks such as Caffe and Torch is a plus.
In addition to a great academic track record, we also require you to have outstanding programming capabilities. You will be working on and extending our well architected and cleanly implemented C++14 computer vision codebase, which serves as a technical foundation for our products. You will be responsible for maintaining the high level of quality of our codebase. We are not seeking candidates who can merely use or have been in contact with the popular computer vision libraries; instead, we are looking for candidates who would easily be able to create better versions of them.
Preferably you also have
- An excellent academic record in computer vision and machine learning, and a strong mathematical mindset.
- Hands-on experience with relevant topics, such as real-time augmented reality and/or with deep learning for large scale recognition.
- Several years of experience with modern C++. You are able to write clean, elegant and maintainable high-performance code, judiciously using the latest C++ 11/14 features.
- A very good understanding of software architecture and programming patterns and paradigms. You know the ins and outs of generic programming and data oriented design. You have experience with writing multithreaded code.
- A strong publication record w.r.t. the relevant top-tier conferences and journals (CVPR, ICCV, ECCV, TPAMI, IJCV, etc.)
- Experience with deep learning frameworks such as Caffe or Torch, as well as with GPU-bound high performance computing
- Prior experience with programming mobile platforms, in particular with exploiting their hardware capabilities optimally, via SIMD code (NEON, SSE) or GPGPU programming.
- Experience with prototyping computer vision algorithms using Python or Lua.