Post-doctoral research position in Computer Vision
Why joinWe work on interesting technical challenges using cutting edge technologies in a high performance, agile organization with an innovative environment & high level of autonomy, putting you in the driver’s seat. We have a team-based structure with professional management & continuous development, where you will be exposed to many areas within Autonomous Driving, maximizing development opportunities.
The position is initially offered for two years, but can be extended. Zenuity and the Computer Vision Group offer a creative international environment and a possibility to conduct competitive research in a very active field worldwide. Involvement in supervision of MSc and PhD students is also foreseen.
Applicants are expected to have finished, or be about to finish their PhD degrees, to have a strong background in Computer Vision or Machine Learning, and to have a track record of publications in top conferences and journals. Swedish language skills are not required, English is mandatory.
Driverless vehicles and advanced safety systems are reliant on being able to perceive, interpret and understand the 3D scene surrounding the vehicle. Sensors such as cameras, laser scanners and radars are important for this purpose. In this project, we are interested in developing new machine learning algorithms for improving the state-of-the-art in 3D scene perception for self-driving cars using the camera as a primary sensor.
In recent years, the performance of computer vision applications, like image classification, object recognition and detection has dramatically improved due to the Deep Learning paradigm. The breakthrough has been enabled by theoretical developments, increased computational power and big annotated datasets. Still, the challenges posed by a fully autonomous vehicle due to varying scene types (urban, rural, countryside), seasons, weather, lighting and traffic conditions are enormous and yet to be solved.
The overall goal of the project is to develop algorithms that are capable of 3D surround scene perception from a moving vehicle equipped with multiple cameras and laser scanners. For instance, we want to develop algorithms that are capable of finding objects of interest (cars, pedestrians, traffic lights etc.) and determine their poses in the 3D scene relative the moving vehicle. Most work in the literature is focused on finding bounding boxes in the image domain using a single image as input. Such approaches are often hampered by the fact that objects are occluding each other.
Another research topic that we intend to explore is the problem of semi-automatic 3D annotation of large-scale datasets. In addition to the object instances as described above, we want to develop algorithms that simultaneously recover geometric surfaces and their semantic labels like road, vegetation, buildings and sky. Simultaneous recovery of both objects and environment may support each other for a consistent scene interpretation, and may facilitate the development of efficient annotation tools.
The post-doctoral research fellow will share her/his time 50-50 between Zenuity´s headquarters and Chalmers University of Technology, both located in Gothenburg, Sweden.
Prof. Fredrik Kahl, Head of the Computer Vision Group, Chalmers University of Technology.
Erik Rosén, Technical Leader of Deep Learning, Zenuity AB.
E-mail: firstname.lastname@example.org. Phone: +46 731 25 80 96. www. zenuity.com
Interviews are held on a continuous basis, so we highly recommend that you submit your application at your earliest convenience.