no code implementations • 20 Apr 2024 • Ganesh Sistu, Senthil Yogamani
To the best of our knowledge, this is the first detailed study on object detection on fisheye cameras for autonomous driving scenarios.
no code implementations • 11 Apr 2024 • Muhammad Adeel Hafeez, Michael G. Madden, Ganesh Sistu, Ihsan Ullah
The optimized loss function is a combination of weighted losses to which enhance robustness and generalization: Mean Absolute Error (MAE), Edge Loss and Structural Similarity Index (SSIM).
1 code implementation • 7 Apr 2024 • Saravanabalagi Ramachandran, Jonathan Horgan, Ganesh Sistu, John McDonald
Based on our empirical analysis of multiple runs, we identify that continuity and distinctiveness are crucial characteristics for an optimal global descriptor that enable efficient and scalable hierarchical mapping, and present a methodology for quantifying and contrasting these characteristics across different global descriptors.
no code implementations • 6 Apr 2024 • Muhammad Asad, Ihsan Ullah, Ganesh Sistu, Michael G. Madden
The first is that we compute the novelty probability by linearizing the manifold that holds the structure of the inlier distribution.
no code implementations • 1 Feb 2024 • Arindam Das, Sudarshan Paul, Niko Scholz, Akhilesh Kumar Malviya, Ganesh Sistu, Ujjwal Bhattacharya, Ciarán Eising
Therefore, we present, to our knowledge, the first end-to-end multimodal fusion model tailored for efficient obstacle perception in a bird's-eye-view (BEV) perspective, utilizing fisheye cameras and ultrasonic sensors.
no code implementations • 31 Dec 2023 • Saravanabalagi Ramachandran, Nathaniel Cibik, Ganesh Sistu, John McDonald
Motion segmentation is a complex yet indispensable task in autonomous driving.
no code implementations • 20 Dec 2023 • Sushil Sharma, Aryan Singh, Ganesh Sistu, Mark Halton, Ciarán Eising
Predicting the trajectory of an ego vehicle is a critical component of autonomous driving systems.
no code implementations • 20 Dec 2023 • Sushil Sharma, Arindam Das, Ganesh Sistu, Mark Halton, Ciarán Eising
The proposed method in this paper predicts trajectories by considering perception and trajectory prediction as a unified system.
no code implementations • 16 Aug 2023 • Ciarán Hogan, Ganesh Sistu, Ciarán Eising
The framework is self-supervised and doesn't require any labelling or supervision to learn the calibration parameters.
1 code implementation • 18 Jul 2023 • Kaavya Rekanar, Ciarán Eising, Ganesh Sistu, Martin Hayes
This short paper presents a preliminary analysis of three popular Visual Question Answering (VQA) models, namely ViLBERT, ViLT, and LXMERT, in the context of answering questions relating to driving scenarios.
1 code implementation • 11 Jul 2023 • Sushil Sharma, Ganesh Sistu, Lucie Yahiaoui, Arindam Das, Mark Halton, Ciarán Eising
To address this limitation, we have developed a synthetic dataset for short-term trajectory prediction tasks using the CARLA simulator.
no code implementations • 14 Apr 2023 • Mena Nagiub, Thorsten Beuth, Ganesh Sistu, Heinrich Gotzig, Ciarán Eising
Indirect Time of Flight LiDARs can indirectly calculate the scene's depth from the phase shift angle between transmitted and received laser signals with amplitudes modulated at a predefined frequency.
no code implementations • 24 Feb 2023 • Arindam Das, Sudip Das, Ganesh Sistu, Jonathan Horgan, Ujjwal Bhattacharya, Edward Jones, Martin Glavin, Ciarán Eising
Multimodal learning, particularly for pedestrian detection, has recently received emphasis due to its capability to function equally well in several critical autonomous driving scenarios such as low-light, night-time, and adverse weather conditions.
no code implementations • 26 Oct 2022 • Saravanabalagi Ramachandran, Jonathan Horgan, Ganesh Sistu, John McDonald
We train a Variational Autoencoder in an unsupervised manner and map images to a constrained multi-dimensional latent space and use the latent vectors as compact embeddings that serve as global descriptors for images.
no code implementations • 26 Jun 2022 • Saravanabalagi Ramachandran, Ganesh Sistu, Varun Ravi Kumar, John McDonald, Senthil Yogamani
Object detection is a comprehensively studied problem in autonomous driving.
no code implementations • 15 Jun 2022 • Arindam Das, Sudip Das, Ganesh Sistu, Jonathan Horgan, Ujjwal Bhattacharya, Edward Jones, Martin Glavin, Ciarán Eising
The proposed framework has improved state-of-the-art performances of pose estimation, pedestrian detection, and instance segmentation.
Ranked #18 on Pose Estimation on COCO test-dev
1 code implementation • 31 May 2022 • Pramit Dutta, Ganesh Sistu, Senthil Yogamani, Edgar Galván, John McDonald
In this paper, we evaluate the use of vision transformers (ViT) as a backbone architecture to generate BEV maps.
no code implementations • 4 May 2022 • Fergal Stapleton, Edgar Galván, Ganesh Sistu, Senthil Yogamani
The incentive for using Evolutionary Algorithms (EAs) for the automated optimization and training of deep neural networks (DNNs), a process referred to as neuroevolution, has gained momentum in recent years.
no code implementations • 17 Jul 2021 • Saravanabalagi Ramachandran, Ganesh Sistu, John McDonald, Senthil Yogamani
This challenge served as a medium to investigate the challenges and new methodologies to handle the complexities with perception on fisheye images.
no code implementations • 17 May 2021 • Michal Uricar, Ganesh Sistu, Lucie Yahiaoui, Senthil Yogamani
Manual annotation of soiling on surround view cameras is a very challenging and expensive task.
no code implementations • 27 Feb 2021 • Anna Konrad, Ciarán Eising, Ganesh Sistu, John McDonald, Rudi Villing, Senthil Yogamani
Keypoint detection and description is a commonly used building block in computer vision systems particularly for robotics and autonomous driving.
1 code implementation • 15 Feb 2021 • Varun Ravi Kumar, Senthil Yogamani, Hazem Rashed, Ganesh Sistu, Christian Witt, Isabelle Leang, Stefan Milz, Patrick Mäder
We obtain the state-of-the-art results on KITTI for depth estimation and pose estimation tasks and competitive performance on the other tasks.
no code implementations • 3 Dec 2020 • Hazem Rashed, Eslam Mohamed, Ganesh Sistu, Varun Ravi Kumar, Ciaran Eising, Ahmad El-Sallab, Senthil Yogamani
It is the first detailed study on object detection on fisheye cameras for autonomous driving scenarios to the best of our knowledge.
1 code implementation • 16 Oct 2020 • Sumanth Chennupati, Venkatraman Narayanan, Ganesh Sistu, Senthil Yogamani, Samir A Rawashdeh
Instance contours along with semantic segmentation yield a boundary aware semantic segmentation of things.
no code implementations • 1 Jul 2020 • Arindam Das, Pavel Krizek, Ganesh Sistu, Fabian Burger, Sankaralingam Madasamy, Michal Uricar, Varun Ravi Kumar, Senthil Yogamani
Localized detection of soiling in an image is necessary to control the cleaning system.
no code implementations • 7 Jan 2020 • Isabelle Leang, Ganesh Sistu, Fabian Burger, Andrei Bursuc, Senthil Yogamani
Deep multi-task networks are of particular interest for autonomous driving systems.
no code implementations • 23 Dec 2019 • Pullarao Maddu, Wayne Doherty, Ganesh Sistu, Isabelle Leang, Michal Uricar, Sumanth Chennupati, Hazem Rashed, Jonathan Horgan, Ciaran Hughes, Senthil Yogamani
We provide a holistic overview of an industrial system covering the embedded system, use cases and the deep learning architecture.
no code implementations • 4 Dec 2019 • Michal Uricar, Ganesh Sistu, Hazem Rashed, Antonin Vobecky, Varun Ravi Kumar, Pavel Krizek, Fabian Burger, Senthil Yogamani
We propose a novel GAN based algorithm for generating unseen patterns of soiled images.
no code implementations • 11 Oct 2019 • Hazem Rashed, Mohamed Ramzy, Victor Vaquero, Ahmad El Sallab, Ganesh Sistu, Senthil Yogamani
In this work, we propose a robust and real-time CNN architecture for Moving Object Detection (MOD) under low-light conditions by capturing motion information from both camera and LiDAR sensors.
no code implementations • 30 Aug 2019 • Marie Yahiaoui, Hazem Rashed, Letizia Mariotti, Ganesh Sistu, Ian Clancy, Lucie Yahiaoui, Varun Ravi Kumar, Senthil Yogamani
In this work, we propose a CNN architecture for moving object detection using fisheye images that were captured in autonomous driving environment.
1 code implementation • ICCV 2019 • Senthil Yogamani, Ciaran Hughes, Jonathan Horgan, Ganesh Sistu, Padraig Varley, Derek O'Dea, Michal Uricar, Stefan Milz, Martin Simon, Karl Amende, Christian Witt, Hazem Rashed, Sumanth Chennupati, Sanjaya Nayak, Saquib Mansoor, Xavier Perroton, Patrick Perez
Fisheye cameras are commonly employed for obtaining a large field of view in surveillance, augmented reality and in particular automotive applications.
no code implementations • 4 May 2019 • Michal Uricar, Pavel Krizek, Ganesh Sistu, Senthil Yogamani
Cameras are an essential part of sensor suite in autonomous driving.
no code implementations • 15 Apr 2019 • Sumanth Chennupati, Ganesh Sistu, Senthil Yogamani, Samir A Rawashdeh
In this work, we propose a multi-stream multi-task network to take advantage of using feature representations from preceding frames in a video sequence for joint learning of segmentation, depth, and motion.
no code implementations • 10 Feb 2019 • Ganesh Sistu, Isabelle Leang, Sumanth Chennupati, Senthil Yogamani, Ciaran Hughes, Stefan Milz, Samir Rawashdeh
In this paper, we propose a joint multi-task network design for learning several tasks simultaneously.
no code implementations • 17 Jan 2019 • Sumanth Chennupati, Ganesh Sistu, Senthil Yogamani, Samir Rawashdeh
Decision making in automated driving is highly specific to the environment and thus semantic segmentation plays a key role in recognizing the objects in the environment around the car.
no code implementations • 12 Jan 2019 • Ganesh Sistu, Isabelle Leang, Senthil Yogamani
In this paper, we present a joint multi-task network design for learning object detection and semantic segmentation simultaneously.
no code implementations • 8 Jan 2019 • Ganesh Sistu, Sumanth Chennupati, Senthil Yogamani
We propose two simple high-level architectures based on Recurrent FCN (RFCN) and Multi-Stream FCN (MSFCN) networks.