Torna alla lista annunci
Judo - Pose Detection Thesis proposal
The initial part of this work will consist of a research in literature on the state of the art for human posture estimation and video understanding algorithms.
The second part will be dedicated to test and compare such algorithms on real datasets, in terms of accuracy and computation time.
The final goal consists in combining the chosen algorithms in a single pipeline, in order to improve the classification accuracy of sport actions, with particular interest in Judo Ne Waza and Nage Waza techniques.
Addfor S.p.A. works on Artificial Intelligence solutions since 2008, pursuing a unique mission: provide the best existing technologies to its customers. We are organized into two divisions: The first division is the R&D department that is responsible for identifying, evaluating and reviewing all the most advanced Artificial Intelligence algorithms made available from world’s leading research centers, universities and companies (ex: Facebook, Google, DeepMind, Open AI, Stanford, etc). Our second division, has an application focus: to industrialize the chosen technologies to meet the specific customers needs. In order to deliver robust end-to-end integrated solution, we have finalized strategic partnerships with leading hardware manufacturing companies and suppliers: VI-Grade, NVIDIA and IBM. But, the most important of all, is that in Addfor we have fun!