Browsing by Author "Grenier, Amélie"

Now showing 1 - 3 of 3
  • Thumbnail Image
    ItemOpen Access
    Towards scene understanding implementing the stixel world
    (IEEE, 2019-03-07) Grenier, Amélie; Alzoubi, Alaa; Feetham, Luke; Nam, David
    In this paper, we present our work towards scene understanding based on modeling the scene prior to understanding its content. We describe the environment representation model used, the Stixel World, and its benefits for compact scene representation. We show our preliminary results of its application in a diverse environment and the limitations reached in our experiments using imaging systems. We argue that this method has been developed in an ideal scenario and does not generalise well to uncommon changes in the environment. We also found that this method is sensitive to the quality of the stereo rectification and the calibration of the optics, among other parameters, which makes it time-consuming and delicate to prepare in real-time applications. We think that pixel-wise semantic segmentation techniques can address some of the shortcomings of the concept presented in a theoretical discussion.
  • Thumbnail Image
    ItemOpen Access
    Visual Scene Understanding for Self-Driving Cars Using Deep Learning and Stereovision
    (Cranfield University, 2019-02-07 16:16) Grenier, Amélie
    Poster presented at the 2018 Defence and Security Doctoral Symposium.Autonomous driving has been rapidly evolving for the last few years and there is a lot of fervour in increasing the intelligence of these vehicles. One key aspect of a self-driving car is its ability to sense the environment in order to be aware of its surrounding.Our interest lies in using computer vision and deep learning techniques to detect surrounding entities; localising and recognising them. Here, we present a novel deconvolutional neural network for semantic segmentation, combined with disparity map information to localise each vehicle in front of the ego-vehicle, including occluded instances, in an urban traffic environment. We also compare our approach with state-of-the-art instance segmentation methods. In the future, we will extend our work to other types of obstacles, to improve awareness and increase obstacle avoidance and path finding capabilities of a vehicle.
  • Thumbnail Image
    ItemOpen Access
    What is Where? Visual Understanding for Autonomous Cars
    (Cranfield University, 2020-01-09 11:36) Grenier, Amélie
    Autonomous driving has been rapidly evolving for the last few years and there is a lot of fervour in increasing the intelligence of these vehicles. One key aspect of a self-driving car is its ability to sense the environment in order to be aware of its surroundings and consecutively take better decisions.While the right combination of sensors is widely debated, my research interest lies in using computer vision and machine learning techniques to detect, localise and recognise surrounding entities. My talk will describe my research objectives and the expected outcome. It will address some of the encountered challenges, resulting from the urban traffic environment context and my sensor choice. It will include a mention of the algorithms that I have tested so far and those currently in development. You will have a glance at some of the questions that researchers are presently trying to answer in this interdisciplinary field.