17 juin 2019
Sciences Informatiques : Les résultats de la recherche académique financée par l'IMRA presentés à CVPR 2019
At the CVPR 2019 conference, IMRA's collaborators from the Computer Vision Group at the University of Freiburg, Germany presented the results of their work on multiple future prediction funded by IMRA Europe S.A.S.
The algorithm is able to predict that the cyclist arriving to a roundabout (inside the red boxes) can follow three different paths, each one with its own probability of occurrence (as illustrated with different heatmaps).
Last June at the CVPR'19 conference at LOng Beach, California, our collaborators from the Computer Vision Group at the University of Freiburg, Germany presented the results of their work on multiple future prediction funded by IMRA Europe S.A.S. This work is entitled "Overcoming Limitations of Mixture Density Networks: A sampling and Fitting Framework for Multimodal Future Prediction" and presents a novel approach for predicting multiple future positions of a moving object as multimodal probability distributions. Congratulations to Osama Makansi, the main author of this work, and to all his colleagues, for this great academic achievement !
Article and related material: here (CVPR 2019 at CVF)
Video presentation: here
Last June at the CVPR'19 conference at LOng Beach, California, our collaborators from the Computer Vision Group at the University of Freiburg, Germany presented the results of their work on multiple future prediction funded by IMRA Europe S.A.S. This work is entitled "Overcoming Limitations of Mixture Density Networks: A sampling and Fitting Framework for Multimodal Future Prediction" and presents a novel approach for predicting multiple future positions of a moving object as multimodal probability distributions. Congratulations to Osama Makansi, the main author of this work, and to all his colleagues, for this great academic achievement !
Article and related material: here (CVPR 2019 at CVF)
Video presentation: here