SURVISION, world leader in vehicle identification video sensors, based in the Paris Region and in the United States, is looking for a deep learning R & D Engineer.
In a small, young and dynamic team, you design innovative algorithms in the field of video analysis and deep learning applied to the detection, tracking and identification of vehicles.
You contribute to defining the architecture and the developments of the new concepts by taking into account the technical constraints related to the platforms used in our products.
You study the models of detection of objects most adapted to the studied problems (CNN, auto-encoders, RNN, LSTM, Q-learning, etc.).
You implement these innovative approaches by developing demonstrators.
You then participate in the integration and optimization phases of these architectures to achieve real-time performance on embedded applications (memory and power constraints).
You also contribute to the construction of acquisition and learning data acquisition architectures.
Graduate (e) of a large engineering school or holder of a PhD thesis, you justify a first experience in the field of machine learning applied to image processing or video.
You have skills in image and video processing algorithms, Deep Learning concepts and architectures and C / C ++, Python programming ... Knowledge of a framework such as TensorFlow or PyTorch is a prerequisite.
You may have skills in image sensors.
You are recognized for your team spirit, initiative and curiosity.
You have a level in English allowing you to easily exchange daily in a professional environment.
Please note: the original job description is in French