Filters








7 Hits in 0.82 sec

Point Cloud based Hierarchical Deep Odometry Estimation [article]

Farzan Erlik Nowruzi, Dhanvin Kolhatkar, Prince Kapoor, Robert Laganiere
2021 arXiv   pre-print
Inspired by a hierarchical homography network (Nowruzi et al., 2017) , we use multiple layers of the same network to train on the residuals of previous predictions.  ... 
arXiv:2103.03394v1 fatcat:r4k5xqahqfe4di5b4atdrhczue

Deep Open Space Segmentation using Automotive Radar [article]

Farzan Erlik Nowruzi, Dhanvin Kolhatkar, Prince Kapoor, Fahed Al Hassanat, Elnaz Jahani Heravi, Robert Laganiere, Julien Rebut, Waqas Malik
2020 arXiv   pre-print
In this work, we propose the use of radar with advanced deep segmentation models to identify open space in parking scenarios. A publically available dataset of radar observations called SCORP was collected. Deep models are evaluated with various radar input representations. Our proposed approach achieves low memory usage and real-time processing speeds, and is thus very well suited for embedded deployment.
arXiv:2004.03449v1 fatcat:by37fu7uanesnc2gl33xfpcdbm

Homography Estimation from Image Pairs with Hierarchical Convolutional Networks

Nathalie Japkowicz, Farzan Erlik Nowruzi, Robert Laganiere
2017 2017 IEEE International Conference on Computer Vision Workshops (ICCVW)  
In this paper, we introduce a hierarchy of twin convolutional regression networks to estimate the homography between a pair of images. In this framework, networks are stacked sequentially in order to reduce error bounds of the estimate. At every convolutional network module, features from each image are extracted independently, given a shared set of kernels, also known as Siamese network model. Later on in the process, they are merged together to estimate the homography. Further, we evaluate
more » ... compare effects of various training parameters in this context. We show that given the iterative nature of the framework, highly complicated models are not necessarily required, and high performance is achieved via hierarchical arrangement of simple models. Effectiveness of the proposed method is shown through experiments on MSCOCO dataset, in which it significantly outperforms the state-of-the-art.
doi:10.1109/iccvw.2017.111 dblp:conf/iccvw/JapkowiczNL17 fatcat:h4tijvg6t5fcdgsjjnicn423vi

PolarNet: Accelerated Deep Open Space Segmentation Using Automotive Radar in Polar Domain [article]

Farzan Erlik Nowruzi, Dhanvin Kolhatkar, Prince Kapoor, Elnaz Jahani Heravi, Fahed Al Hassanat, Robert Laganiere, Julien Rebut, Waqas Malik
2021 arXiv   pre-print
Most recently, (Nowruzi et al., 2020) introduced a novel dataset for open space segmentation in automotive parking scenarios.  ...  ., 2019) (Nowruzi et al., 2020) that have recently been introduced to enable the scientific community to expand the boundaries of knowledge in this field.  ... 
arXiv:2103.03387v1 fatcat:mwdttkalanbwxeypnp6x2iseva

In-Vehicle Occupancy Detection With Convolutional Networks on Thermal Images

Farzan Erlik Nowruzi, Wassim A. El Ahmar, Robert Laganiere, Amir H. Ghods
2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
Counting people is a growing field of interest for researchers in recent years. In-vehicle passenger counting is an interesting problem in this domain that has several applications including High Occupancy Vehicle (HOV) lanes. In this paper, present a new in-vehicle thermal image dataset. We propose a tiny convolutional model to count on-board passengers and compare it to well known methods. We show that our model surpasses state-of-the-art methods in classification and has comparable
more » ... e in detection. Moreover, our model outperforms the state-of-the-art architectures in terms of speed, making it suitable for deployment on embedded platforms. We present the results of multiple deep learning models and thoroughly analyze them.
doi:10.1109/cvprw.2019.00124 dblp:conf/cvpr/NowruziALG19 fatcat:jwbe5y5gezgqdmfrvc6du3vnsy

Fast Human Head and Shoulder Detection Using Convolutional Networks and RGBD Data

Wassim A. El Ahmar, Farzan Erlik Nowruzi, Robert Laganiere
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
We introduce a new real-time approach for human head and shoulder detection from RGB-D data based on a combination of image processing and deep learning approaches. Candidate head-top locations (CHL) are generated from a fast and accurate image processing algorithm that operates on depth data. We propose enhancements to the CHL algorithm making it three times faster. Various deep learning models are then evaluated for the tasks of classification and detection on the candidate head-top locations
more » ... to regress the head bounding boxes and detect shoulder keypoints. We propose three different models based on convolutional neural networks for this problem. Experimental results for different architectures of our model are discussed. We also compare the performance of our models to other state of the art methods in terms of accuracy of detections and computational cost and show that our proposed models are on par with the state of the art in terms of precision-recall of head detection and precision of shoulders detection, with the biggest advantage of our models being in terms of computation time. We also analyze the effect of adding the depth channel on the performance of the network.
doi:10.1109/cvprw50498.2020.00061 dblp:conf/cvpr/AhmarNL20 fatcat:p4rlupyymfb27gpiyehf2msr74

Table of Contents

2019 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
Erlik Nowruzi (missing), Wassim A.  ...  Guoliang Fan (missing), Guangfeng Lin (missing), Wanjun Chen (missing), Xiaorong Pan (missing), and Hong Zhu (missing) In-Vehicle Occupancy Detection With Convolutional Networks on Thermal Images 941 Farzan  ... 
doi:10.1109/cvprw.2019.00004 fatcat:h7xpqwyrofdxniqtxbodn66mpy