Pub Date : 2023-01-16DOI: 10.2352/ei.2023.35.16.avm-116
T. Brophy, B. Deegan, J. Salado, A. Tena, Patrick Denny, M. Glavin, Enda Ward, J. Horgan, E. Jones
{"title":"The design and validation of a rain model for a simulated automotive environment","authors":"T. Brophy, B. Deegan, J. Salado, A. Tena, Patrick Denny, M. Glavin, Enda Ward, J. Horgan, E. Jones","doi":"10.2352/ei.2023.35.16.avm-116","DOIUrl":"https://doi.org/10.2352/ei.2023.35.16.avm-116","url":null,"abstract":"","PeriodicalId":177462,"journal":{"name":"Autonomous Vehicles and Machines","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128839484","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-02-26DOI: 10.2352/EI.2023.35.16.AVM-122
Chen-Chou Lo, P. Vandewalle
Recently, several works have proposed fusing radar data as an additional perceptual signal into monocular depth estimation models because radar data is robust against varying light and weather conditions. Although improved performances were reported in prior works, it is still hard to tell how much depth information radar can contribute to a depth estimation model. In this paper, we propose radar inference and supervision experiments to investigate the intrinsic depth potential of radar data using state-of-the-art depth estimation models on the nuScenes dataset. In the inference experiment, the model predicts depth by taking only radar as input to demonstrate the inference capability using radar data. In the supervision experiment, a monocular depth estimation model is trained under radar supervision to show the intrinsic depth information that radar can contribute. Our experiments demonstrate that the model using only sparse radar as input can detect the shape of surroundings to a certain extent in the predicted depth. Furthermore, the monocular depth estimation model supervised by preprocessed radar achieves a good performance compared to the baseline model trained with sparse lidar supervision.
{"title":"How much depth information can radar contribute to a depth estimation model?","authors":"Chen-Chou Lo, P. Vandewalle","doi":"10.2352/EI.2023.35.16.AVM-122","DOIUrl":"https://doi.org/10.2352/EI.2023.35.16.AVM-122","url":null,"abstract":"Recently, several works have proposed fusing radar data as an additional perceptual signal into monocular depth estimation models because radar data is robust against varying light and weather conditions. Although improved performances were reported in prior works, it is still hard to tell how much depth information radar can contribute to a depth estimation model. In this paper, we propose radar inference and supervision experiments to investigate the intrinsic depth potential of radar data using state-of-the-art depth estimation models on the nuScenes dataset. In the inference experiment, the model predicts depth by taking only radar as input to demonstrate the inference capability using radar data. In the supervision experiment, a monocular depth estimation model is trained under radar supervision to show the intrinsic depth information that radar can contribute. Our experiments demonstrate that the model using only sparse radar as input can detect the shape of surroundings to a certain extent in the predicted depth. Furthermore, the monocular depth estimation model supervised by preprocessed radar achieves a good performance compared to the baseline model trained with sparse lidar supervision.","PeriodicalId":177462,"journal":{"name":"Autonomous Vehicles and Machines","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-02-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121519799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-01-16DOI: 10.2352/ei.2022.34.16.avm-109
O. Skorka, Paul Romanczyk
The IEEE P2020 Noise standard is built upon methodology that is discussed by other photography and camera standards. It includes extensions and adjustments to support operating modes and conditions that are relevant to automotive cameras. This work presents methods and procedures that are covered by the IEEE P2020 Noise standard to derive sensor-level and camera-level noise image quality factors from dark statistics, photon-transfer and signal-to-noise ratio curves, and signal falloff. Example implementations and experimental results are shown from work that was done with automotive cameras which were activated and tested under conditions that are relevant to automotive applications.
{"title":"A review of IEEE P2020 noise metrics","authors":"O. Skorka, Paul Romanczyk","doi":"10.2352/ei.2022.34.16.avm-109","DOIUrl":"https://doi.org/10.2352/ei.2022.34.16.avm-109","url":null,"abstract":"The IEEE P2020 Noise standard is built upon methodology that is discussed by other photography and camera standards. It includes extensions and adjustments to support operating modes and conditions that are relevant to automotive cameras. This work presents methods and procedures that are covered by the IEEE P2020 Noise standard to derive sensor-level and camera-level noise image quality factors from dark statistics, photon-transfer and signal-to-noise ratio curves, and signal falloff. Example implementations and experimental results are shown from work that was done with automotive cameras which were activated and tested under conditions that are relevant to automotive applications.","PeriodicalId":177462,"journal":{"name":"Autonomous Vehicles and Machines","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125987041","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-01-16DOI: 10.2352/ei.2022.34.16.avm-101
Mattis Brummel, Patrick Müller, Alexander Braun
Since it is essential for Computer Vision systems to reliably perform in safety-critical applications such as autonomous vehicles, there is a need to evaluate their robustness to naturally occurring image perturbations. More specifically, the performance of Computer Vision systems needs to be linked to the image quality, which hasn’t received much research attention so far. In fact, aberrations of a camera system are always spatially variable over the Field of View, which may influence the performance of Computer Vision systems dependent on the degree of local aberrations. Therefore, the goal is to evaluate the performance of Computer Vision systems under effects of defocus by taking into account the spatial domain. Large-scale Autonomous Driving datasets are degraded by a parameterized optical model to simulate driving scenes under physically realistic effects of defocus. Using standard evaluation metrics, the Spatial Recall Index (SRI) and the new Spatial Precision Index (SPI), the performance of Computer Visions systems on these degraded datasets are compared with the optical performance of the applied optical model. A correlation could be observed between the spatially varying optical performance and the spatial performance of Instance Segmentation systems.
{"title":"Spatial precision and recall indices to assess the performance of instance segmentation algorithms","authors":"Mattis Brummel, Patrick Müller, Alexander Braun","doi":"10.2352/ei.2022.34.16.avm-101","DOIUrl":"https://doi.org/10.2352/ei.2022.34.16.avm-101","url":null,"abstract":"Since it is essential for Computer Vision systems to reliably perform in safety-critical applications such as autonomous vehicles, there is a need to evaluate their robustness to naturally occurring image perturbations. More specifically, the performance of Computer Vision systems needs to be linked to the image quality, which hasn’t received much research attention so far. In fact, aberrations of a camera system are always spatially variable over the Field of View, which may influence the performance of Computer Vision systems dependent on the degree of local aberrations. Therefore, the goal is to evaluate the performance of Computer Vision systems under effects of defocus by taking into account the spatial domain. Large-scale Autonomous Driving datasets are degraded by a parameterized optical model to simulate driving scenes under physically realistic effects of defocus. Using standard evaluation metrics, the Spatial Recall Index (SRI) and the new Spatial Precision Index (SPI), the performance of Computer Visions systems on these degraded datasets are compared with the optical performance of the applied optical model. A correlation could be observed between the spatially varying optical performance and the spatial performance of Instance Segmentation systems.","PeriodicalId":177462,"journal":{"name":"Autonomous Vehicles and Machines","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2022-01-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115265319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-07-15DOI: 10.2352/j.imagingsci.technol.2021.65.6.060408
Ibrahim Sobh, Ahmed Hamed, V. Kumar, S. Yogamani
In recent years, deep neural networks (DNNs) have accomplished impressive success in various applications, including autonomous driving perception tasks. However, current deep neural networks are easily deceived by adversarial attacks. This vulnerability raises significant concerns, particularly in safety-critical applications. As a result, research into attacking and defending DNNs has gained much coverage. In this work, detailed adversarial attacks are applied on a diverse multi-task visual perception deep network across distance estimation, semantic segmentation, motion detection, and object detection. The experiments consider both white and black box attacks for targeted and un-targeted cases, while attacking a task and inspecting the effect on all others, in addition to inspecting the effect of applying a simple defense method. We conclude this paper by comparing and discussing the experimental results, proposing insights and future work. The visualizations of the attacks are available at https://youtu.be/6AixN90budY.
{"title":"Adversarial Attacks on Multi-task Visual Perception for Autonomous Driving","authors":"Ibrahim Sobh, Ahmed Hamed, V. Kumar, S. Yogamani","doi":"10.2352/j.imagingsci.technol.2021.65.6.060408","DOIUrl":"https://doi.org/10.2352/j.imagingsci.technol.2021.65.6.060408","url":null,"abstract":"\u0000 In recent years, deep neural networks (DNNs) have accomplished impressive success in various applications, including autonomous driving perception tasks. However, current deep neural networks are easily deceived by adversarial attacks. This vulnerability raises significant concerns, particularly in safety-critical applications. As a result, research into attacking and defending DNNs has gained much coverage. In this work, detailed adversarial attacks are applied on a diverse multi-task visual perception deep network across distance estimation, semantic segmentation, motion detection, and object detection. The experiments consider both white and black box attacks for targeted and un-targeted cases, while attacking a task and inspecting the effect on all others, in addition to inspecting the effect of applying a simple defense method. We conclude this paper by comparing and discussing the experimental results, proposing insights and future work. The visualizations of the attacks are available at https://youtu.be/6AixN90budY.\u0000","PeriodicalId":177462,"journal":{"name":"Autonomous Vehicles and Machines","volume":"98 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126090231","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-01-18DOI: 10.2352/issn.2470-1173.2021.17.avm-113
Mihir Mody, Kedar Chitnis, H. Hariyani, Shyam Jagannathan, Jason Jones, G. Shurtz, Abhishek Shankar, Ankur, Mayank Mangla, Sriramakrishnan Govindarajan, Aish Dubey, K. Chirca
Auto-Valet parking is a key emerging function for Advanced Driver Assistance Systems (ADAS) enhancing traditional surround view system providing more autonomy during parking scenario. Auto-Valet parking system is typically built using multiple HW components e.g. ISP, micro-controllers, FPGAs, GPU, Ethernet/PCIe switch etc. Texas Instrument’s new Jacinto7 platform is one of industry’s highest integrated SoC replacing these components with a single TDA4VMID chip. The TDA4VMID SoC can concurrently do analytics (traditional computer vision as well as deep learning) and sophisticated 3D surround view, making it a cost effective and power optimized solution. TDA4VMID is a truly heterogeneous architecture and it can be programmed using an efficient and easy to use OpenVX based middle-ware framework to realize distribution of software components across cores. This paper explains typical functions for analytics and 3D surround view in auto-valet parking system with data-flow and its mapping to multiple cores of TDA4VMID SoC. Auto-valet parking system can be realized on TDA4VMID SOC with complete processing offloaded of host ARM to the rest of SoC cores, providing ample headroom for customers for future proofing as well as ability to add customer specific differentiation.
{"title":"Single Chip Auto-Valet Parking System with TDA4VMID SoC","authors":"Mihir Mody, Kedar Chitnis, H. Hariyani, Shyam Jagannathan, Jason Jones, G. Shurtz, Abhishek Shankar, Ankur, Mayank Mangla, Sriramakrishnan Govindarajan, Aish Dubey, K. Chirca","doi":"10.2352/issn.2470-1173.2021.17.avm-113","DOIUrl":"https://doi.org/10.2352/issn.2470-1173.2021.17.avm-113","url":null,"abstract":"\u0000 Auto-Valet parking is a key emerging function for Advanced Driver Assistance Systems (ADAS) enhancing traditional surround view system providing more autonomy during parking scenario. Auto-Valet parking system is typically built using multiple HW components e.g. ISP, micro-controllers,\u0000 FPGAs, GPU, Ethernet/PCIe switch etc. Texas Instrument’s new Jacinto7 platform is one of industry’s highest integrated SoC replacing these components with a single TDA4VMID chip. The TDA4VMID SoC can concurrently do analytics (traditional computer vision as well as deep learning)\u0000 and sophisticated 3D surround view, making it a cost effective and power optimized solution. TDA4VMID is a truly heterogeneous architecture and it can be programmed using an efficient and easy to use OpenVX based middle-ware framework to realize distribution of software components across cores.\u0000 This paper explains typical functions for analytics and 3D surround view in auto-valet parking system with data-flow and its mapping to multiple cores of TDA4VMID SoC. Auto-valet parking system can be realized on TDA4VMID SOC with complete processing offloaded of host ARM to the rest of SoC\u0000 cores, providing ample headroom for customers for future proofing as well as ability to add customer specific differentiation.\u0000","PeriodicalId":177462,"journal":{"name":"Autonomous Vehicles and Machines","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114578796","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2021-01-18DOI: 10.2352/issn.2470-1173.2021.17.avm-181
V. Kumar, S. Yogamani, Stefan Milz, Patrick Mäder
FisheyeDistanceNet [1] proposed a self-supervised monocular depth estimation method for fisheye cameras with a large field of view (> 180°). To achieve scale-invariant depth estimation, FisheyeDistanceNet supervises depth map predictions over multiple scales during training. To overcome this bottleneck, we incorporate self-attention layers and robust loss function [2] to FisheyeDistanceNet. A general adaptive robust loss function helps obtain sharp depth maps without a need to train over multiple scales and allows us to learn hyperparameters in loss function to aid in better optimization in terms of convergence speed and accuracy. We also ablate the importance of Instance Normalization over Batch Normalization in the network architecture. Finally, we generalize the network to be invariant to camera views by training multiple perspectives using front, rear, and side cameras. Proposed algorithm improvements, FisheyeDistanceNet++, result in 30% relative improvement in RMSE while reducing the training time by 25% on the WoodScape dataset. We also obtain state-of-the-art results on the KITTI dataset, in comparison to other self-supervised monocular methods.
{"title":"FisheyeDistanceNet++: Self-Supervised Fisheye Distance Estimation with Self-Attention, Robust Loss Function and Camera View Generalization","authors":"V. Kumar, S. Yogamani, Stefan Milz, Patrick Mäder","doi":"10.2352/issn.2470-1173.2021.17.avm-181","DOIUrl":"https://doi.org/10.2352/issn.2470-1173.2021.17.avm-181","url":null,"abstract":"\u0000 FisheyeDistanceNet [1] proposed a self-supervised monocular depth estimation method for fisheye cameras with a large field of view (> 180°). To achieve scale-invariant depth estimation, FisheyeDistanceNet supervises depth map predictions over multiple scales during training.\u0000 To overcome this bottleneck, we incorporate self-attention layers and robust loss function [2] to FisheyeDistanceNet. A general adaptive robust loss function helps obtain sharp depth maps without a need to train over multiple scales and allows us to learn hyperparameters in loss function to\u0000 aid in better optimization in terms of convergence speed and accuracy. We also ablate the importance of Instance Normalization over Batch Normalization in the network architecture. Finally, we generalize the network to be invariant to camera views by training multiple perspectives using front,\u0000 rear, and side cameras. Proposed algorithm improvements, FisheyeDistanceNet++, result in 30% relative improvement in RMSE while reducing the training time by 25% on the WoodScape dataset. We also obtain state-of-the-art results on the KITTI dataset, in comparison to other self-supervised monocular\u0000 methods.\u0000","PeriodicalId":177462,"journal":{"name":"Autonomous Vehicles and Machines","volume":"125 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-01-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130618135","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：