Pub Date : 2015-08-27DOI: 10.1109/IVS.2015.7225684
J. Suhr, H. Jung
This paper proposes a fast method for detecting symbolic road markings (SRMs) and stop-lines. The proposed method efficiently restricts the search area based on the lane detection results and finds SRMs and stop-lines in a cost-effective manner. The SRM detector generates multiple SRM candidates using a top-hat filter and projection histogram and classifies their types using a histogram of oriented gradient (HOG) feature and total error rate (TER)-based classifier. The stop-line detector creates stop-line candidates via random sample consensus (RANSAC)-based parallel line pair estimation and verifies them using the HOG feature and TER-based classifier. The proposed method achieves reasonable detection rates and extremely low false positive rates along with a fast computing time.
{"title":"Fast symbolic road marking and stop-line detection for vehicle localization","authors":"J. Suhr, H. Jung","doi":"10.1109/IVS.2015.7225684","DOIUrl":"https://doi.org/10.1109/IVS.2015.7225684","url":null,"abstract":"This paper proposes a fast method for detecting symbolic road markings (SRMs) and stop-lines. The proposed method efficiently restricts the search area based on the lane detection results and finds SRMs and stop-lines in a cost-effective manner. The SRM detector generates multiple SRM candidates using a top-hat filter and projection histogram and classifies their types using a histogram of oriented gradient (HOG) feature and total error rate (TER)-based classifier. The stop-line detector creates stop-line candidates via random sample consensus (RANSAC)-based parallel line pair estimation and verifies them using the HOG feature and TER-based classifier. The proposed method achieves reasonable detection rates and extremely low false positive rates along with a fast computing time.","PeriodicalId":294701,"journal":{"name":"2015 IEEE Intelligent Vehicles Symposium (IV)","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"121377002","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 : 2015-08-27DOI: 10.1109/IVS.2015.7225758
S. Khastgir, S. Birrell, G. Dhadyalla, P. Jennings
Recently there has been a growth in the incorporation of autonomous features within vehicles. From being perceived as a comfort feature, autonomous features in vehicles have now become a safety feature which are foreseen to reduce accidents. This has led to a new trend within the automotive industry of focussing on autonomous features for driver safety, which might ultimately lead to fully autonomous vehicles. Considering the fact that most of the accidents on UK roads occur due to driver error, driver-less vehicles would prove to be a benefit. However with automation, an even greater challenge of system validation in all scenarios needs to be addressed. For this, various methods of validation have been developed by different research organizations and manufacturers, but a standardized process still evades the industry. Some of the existing methods have been discussed in this paper to critically compare their quality of results and ease of execution. Subsequently, a new test platform has been proposed using the 3xD driving simulator which encompasses most requirements of a general testing method. A standardized process which would benefit the industry both in terms of reducing costs of having varied processes, and by increasing customer confidence can be developed using a non-invasive platform like the 3xD driving simulator. The novelty of the 3xD simulator is the ability to drive-in any vehicle (production/prototype) and develop testing methodologies in an immersive wireless environment.
{"title":"Identifying a gap in existing validation methodologies for intelligent automotive systems: Introducing the 3xD simulator","authors":"S. Khastgir, S. Birrell, G. Dhadyalla, P. Jennings","doi":"10.1109/IVS.2015.7225758","DOIUrl":"https://doi.org/10.1109/IVS.2015.7225758","url":null,"abstract":"Recently there has been a growth in the incorporation of autonomous features within vehicles. From being perceived as a comfort feature, autonomous features in vehicles have now become a safety feature which are foreseen to reduce accidents. This has led to a new trend within the automotive industry of focussing on autonomous features for driver safety, which might ultimately lead to fully autonomous vehicles. Considering the fact that most of the accidents on UK roads occur due to driver error, driver-less vehicles would prove to be a benefit. However with automation, an even greater challenge of system validation in all scenarios needs to be addressed. For this, various methods of validation have been developed by different research organizations and manufacturers, but a standardized process still evades the industry. Some of the existing methods have been discussed in this paper to critically compare their quality of results and ease of execution. Subsequently, a new test platform has been proposed using the 3xD driving simulator which encompasses most requirements of a general testing method. A standardized process which would benefit the industry both in terms of reducing costs of having varied processes, and by increasing customer confidence can be developed using a non-invasive platform like the 3xD driving simulator. The novelty of the 3xD simulator is the ability to drive-in any vehicle (production/prototype) and develop testing methodologies in an immersive wireless environment.","PeriodicalId":294701,"journal":{"name":"2015 IEEE Intelligent Vehicles Symposium (IV)","volume":"34 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125101191","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 : 2015-08-27DOI: 10.1109/IVS.2015.7225896
Teng Teng, Luzheng Bi, Xinan Fan
This paper proposes a novel method to recognize an emergency situation by translating EEG signals of a disabled driver while he or she uses a brain-machine interface without using his or her limbs to drive a vehicle. EEG signals were first filtered by independent component analysis along with information entropy. And then the sums of powers of theta wave in the power spectrum of EEG signals from 13 channels were used as features of the classifier built by linear discriminant analysis. The pilot experimental results from two participants in a driving simulator indicated that the model recognized emergency situations (e.g., pedestrian sudden occurrence) 400 ms earlier than the response of drivers with a hit rate of 76.4%, suggesting that the proposed method is feasible. The proposed method can be used as a complementary method to the existing ones based on detecting external objects with sensors.
{"title":"Using EEG to recognize emergency situations for brain-controlled vehicles","authors":"Teng Teng, Luzheng Bi, Xinan Fan","doi":"10.1109/IVS.2015.7225896","DOIUrl":"https://doi.org/10.1109/IVS.2015.7225896","url":null,"abstract":"This paper proposes a novel method to recognize an emergency situation by translating EEG signals of a disabled driver while he or she uses a brain-machine interface without using his or her limbs to drive a vehicle. EEG signals were first filtered by independent component analysis along with information entropy. And then the sums of powers of theta wave in the power spectrum of EEG signals from 13 channels were used as features of the classifier built by linear discriminant analysis. The pilot experimental results from two participants in a driving simulator indicated that the model recognized emergency situations (e.g., pedestrian sudden occurrence) 400 ms earlier than the response of drivers with a hit rate of 76.4%, suggesting that the proposed method is feasible. The proposed method can be used as a complementary method to the existing ones based on detecting external objects with sensors.","PeriodicalId":294701,"journal":{"name":"2015 IEEE Intelligent Vehicles Symposium (IV)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132528378","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 : 2015-08-27DOI: 10.1109/IVS.2015.7225841
Brian K. Mok, Mishel Johns, Key Jung Lee, Hillary Page Ive, D. Miller, Wendy Ju
With automated driving systems, drivers may still be expected to resume full control of the vehicle. While structured transitions where drivers are given warning are desirable, it is critical to benchmark how drivers perform when transition of control is unstructured and occurs without advanced warning. In this study, we observed how participants (N=27) in a driving simulator performed after they were subjected to an emergency loss of automation. We tested three transition time conditions, with an unstructured transition of vehicle control occurring 2 seconds, 5 seconds, or 8 seconds before the participants encountered a road hazard that required the drivers' intervention. Few drivers in the 2 second condition were able to safely negotiate the road hazard situation, while the majority of drivers in 5 or 8 second conditions were able to navigate the hazard safely. Similarly, drivers in 2 second condition rated the vehicle to be less likeable than drivers in 5 and 8 second conditions. From the study results, we are able to narrow in on a minimum amount of time in which drivers can take over the control of vehicle safely and comfortably from the automated system in the advent of an impending road hazard.
{"title":"Timing of unstructured transitions of control in automated driving","authors":"Brian K. Mok, Mishel Johns, Key Jung Lee, Hillary Page Ive, D. Miller, Wendy Ju","doi":"10.1109/IVS.2015.7225841","DOIUrl":"https://doi.org/10.1109/IVS.2015.7225841","url":null,"abstract":"With automated driving systems, drivers may still be expected to resume full control of the vehicle. While structured transitions where drivers are given warning are desirable, it is critical to benchmark how drivers perform when transition of control is unstructured and occurs without advanced warning. In this study, we observed how participants (N=27) in a driving simulator performed after they were subjected to an emergency loss of automation. We tested three transition time conditions, with an unstructured transition of vehicle control occurring 2 seconds, 5 seconds, or 8 seconds before the participants encountered a road hazard that required the drivers' intervention. Few drivers in the 2 second condition were able to safely negotiate the road hazard situation, while the majority of drivers in 5 or 8 second conditions were able to navigate the hazard safely. Similarly, drivers in 2 second condition rated the vehicle to be less likeable than drivers in 5 and 8 second conditions. From the study results, we are able to narrow in on a minimum amount of time in which drivers can take over the control of vehicle safely and comfortably from the automated system in the advent of an impending road hazard.","PeriodicalId":294701,"journal":{"name":"2015 IEEE Intelligent Vehicles Symposium (IV)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133483292","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 : 2015-08-27DOI: 10.1109/IVS.2015.7225744
Jaebum Choi, M. Maurer
Simultaneous localization and mapping (SLAM) plays a significant role in autonomous vehicles when a global navigation satellite system (GNSS) is not available. Environment models and underlying estimation techniques are key factors of this algorithm. In this paper, we present a hybrid map-based SLAM approach using Rao-Blackwellized particle filters (RBPFs). We represent the environment with the hybrid map which consists of feature and grid maps. The joint posterior between the vehicle positions and both maps are maintained using RBPFs. This approach allows a vehicle to update its states in a more robust and efficient way. We derived a novel sampling formula by combining a feature measurement likelihood to the traditional grid-based SLAM framework and can decrease the uncertainty of the predicted vehicle position significantly. Moreover, we represent the grid maps with 3D models because 2D models could be insufficient and less reliable to achieve tasks such as navigation and obstacle avoidance in complex 3D environment. We are also able to show that the 3D grid measurement likelihood has a lower variance and with that we can improve the overall performance of the algorithm.
{"title":"Simultaneous localization and mapping based on the local volumetric hybrid map","authors":"Jaebum Choi, M. Maurer","doi":"10.1109/IVS.2015.7225744","DOIUrl":"https://doi.org/10.1109/IVS.2015.7225744","url":null,"abstract":"Simultaneous localization and mapping (SLAM) plays a significant role in autonomous vehicles when a global navigation satellite system (GNSS) is not available. Environment models and underlying estimation techniques are key factors of this algorithm. In this paper, we present a hybrid map-based SLAM approach using Rao-Blackwellized particle filters (RBPFs). We represent the environment with the hybrid map which consists of feature and grid maps. The joint posterior between the vehicle positions and both maps are maintained using RBPFs. This approach allows a vehicle to update its states in a more robust and efficient way. We derived a novel sampling formula by combining a feature measurement likelihood to the traditional grid-based SLAM framework and can decrease the uncertainty of the predicted vehicle position significantly. Moreover, we represent the grid maps with 3D models because 2D models could be insufficient and less reliable to achieve tasks such as navigation and obstacle avoidance in complex 3D environment. We are also able to show that the 3D grid measurement likelihood has a lower variance and with that we can improve the overall performance of the algorithm.","PeriodicalId":294701,"journal":{"name":"2015 IEEE Intelligent Vehicles Symposium (IV)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133909439","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 : 2015-08-27DOI: 10.1109/IVS.2015.7225766
Panrang Qu, Jianru Xue, Liang Ma, Chao Ma
The Vector Field Histogram (VFH) is a classical motion planning algorithm which is widely used to handle the trajectory planning problem of mobile robots. However, the traditional VFH algorithm is rarely applied to autonomous vehicles due to the vehicle's well-known non-holonomic constraints, especially in urban environments. To address this problem, we propose a constrained VFH algorithm which takes both kinematic and dynamic constraints of the vehicle into consideration. The goal is achieved via two contributions that concern both kinematic and dynamic constraints of the vehicle. First, we develop a new active region for VFH to guarantee that all states within the region are reachable for the vehicle. Second, we improve the cost function to guide the search to favor feasible motion direction for the vehicle. The proposed algorithm is extensively tested in various simulated urban environments, and experimental results validate its efficiency.
矢量场直方图(Vector Field Histogram, VFH)是一种经典的运动规划算法，被广泛用于处理移动机器人的轨迹规划问题。然而，由于众所周知的车辆非完整约束，特别是在城市环境中，传统的VFH算法很少应用于自动驾驶汽车。为了解决这一问题，我们提出了一种同时考虑车辆运动学和动力学约束的约束VFH算法。该目标是通过两个贡献来实现的，这两个贡献涉及车辆的运动学和动力学约束。首先，我们开发了一个新的VFH活动区域，以保证车辆在该区域内的所有状态都是可达的。其次，对代价函数进行改进，使搜索过程更有利于车辆的可行运动方向。在各种模拟城市环境中对该算法进行了广泛的测试，实验结果验证了该算法的有效性。
{"title":"A constrained VFH algorithm for motion planning of autonomous vehicles","authors":"Panrang Qu, Jianru Xue, Liang Ma, Chao Ma","doi":"10.1109/IVS.2015.7225766","DOIUrl":"https://doi.org/10.1109/IVS.2015.7225766","url":null,"abstract":"The Vector Field Histogram (VFH) is a classical motion planning algorithm which is widely used to handle the trajectory planning problem of mobile robots. However, the traditional VFH algorithm is rarely applied to autonomous vehicles due to the vehicle's well-known non-holonomic constraints, especially in urban environments. To address this problem, we propose a constrained VFH algorithm which takes both kinematic and dynamic constraints of the vehicle into consideration. The goal is achieved via two contributions that concern both kinematic and dynamic constraints of the vehicle. First, we develop a new active region for VFH to guarantee that all states within the region are reachable for the vehicle. Second, we improve the cost function to guide the search to favor feasible motion direction for the vehicle. The proposed algorithm is extensively tested in various simulated urban environments, and experimental results validate its efficiency.","PeriodicalId":294701,"journal":{"name":"2015 IEEE Intelligent Vehicles Symposium (IV)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124057862","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 : 2015-08-27DOI: 10.1109/IVS.2015.7225796
Taewoo Kim, Jaewan Lee, K. Yi
This paper presents the maximum tire-road friction coefficient estimation algorithm which considers about the effect of states. Tire force information is an important factor for active safety system. However, it is difficult to estimate due to the dependency on many states such as vehicle speed, tire pressure, and tire wear. In this paper, several experimental researches about the effect of states on the maximum friction coefficient and previous maximum tire-road friction coefficient estimation algorithms are reviewed and summarized. The influential states and the estimation method which doesn't require extra sensors were determined and combined. The proposed algorithm consists of two parts: an interacting multiple models (IMM) based maximum tire-road friction coefficient estimation and an updating sequence based on the effect of vehicle speed. To validate the algorithm, the closed-loop simulation with the advanced emergency braking system (AEBS) has been conducted. It has been shown that the proposed estimation algorithm could enhance the performance of AEBS algorithm.
{"title":"Enhanced maximum tire-road friction coefficient estimation based advanced emergency braking algorithm","authors":"Taewoo Kim, Jaewan Lee, K. Yi","doi":"10.1109/IVS.2015.7225796","DOIUrl":"https://doi.org/10.1109/IVS.2015.7225796","url":null,"abstract":"This paper presents the maximum tire-road friction coefficient estimation algorithm which considers about the effect of states. Tire force information is an important factor for active safety system. However, it is difficult to estimate due to the dependency on many states such as vehicle speed, tire pressure, and tire wear. In this paper, several experimental researches about the effect of states on the maximum friction coefficient and previous maximum tire-road friction coefficient estimation algorithms are reviewed and summarized. The influential states and the estimation method which doesn't require extra sensors were determined and combined. The proposed algorithm consists of two parts: an interacting multiple models (IMM) based maximum tire-road friction coefficient estimation and an updating sequence based on the effect of vehicle speed. To validate the algorithm, the closed-loop simulation with the advanced emergency braking system (AEBS) has been conducted. It has been shown that the proposed estimation algorithm could enhance the performance of AEBS algorithm.","PeriodicalId":294701,"journal":{"name":"2015 IEEE Intelligent Vehicles Symposium (IV)","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127816196","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 : 2015-08-27DOI: 10.1109/IVS.2015.7225904
Donghoon Shin, K. Yi
For the generic assessment and the total management of collision risks in urban driving situations, it is important to estimate and represent the target vehicles' behavior such as yaw rate, absolute velocity and acceleration which are state of the target vehicle. To achieve this, this paper presents a compensation of wireless communication delay for integrated risk management of automated vehicle. Recent developments in vehicle onboard computers and wireless communications devices, also known as dedicated short-range communication (DSRC) devices allow the exchange of information between vehicles (inter-vehicle communications). In an application of vehicle to vehicle (V2V) communication, the most important issue is to handle delay which has a negative impact on safety issue since communication networks generally introduce delays. To cope with this problem, the inter-vehicle communication system is firstly modelled by reflecting signal characteristic. To compensate the communication delay, state augmented estimation algorithm is used based on extended Kalman filters (EKF). The performance of the proposed estimation algorithm is verified via real time simulations.
{"title":"Compensation of wireless communication delay for integrated risk management of automated vehicle","authors":"Donghoon Shin, K. Yi","doi":"10.1109/IVS.2015.7225904","DOIUrl":"https://doi.org/10.1109/IVS.2015.7225904","url":null,"abstract":"For the generic assessment and the total management of collision risks in urban driving situations, it is important to estimate and represent the target vehicles' behavior such as yaw rate, absolute velocity and acceleration which are state of the target vehicle. To achieve this, this paper presents a compensation of wireless communication delay for integrated risk management of automated vehicle. Recent developments in vehicle onboard computers and wireless communications devices, also known as dedicated short-range communication (DSRC) devices allow the exchange of information between vehicles (inter-vehicle communications). In an application of vehicle to vehicle (V2V) communication, the most important issue is to handle delay which has a negative impact on safety issue since communication networks generally introduce delays. To cope with this problem, the inter-vehicle communication system is firstly modelled by reflecting signal characteristic. To compensate the communication delay, state augmented estimation algorithm is used based on extended Kalman filters (EKF). The performance of the proposed estimation algorithm is verified via real time simulations.","PeriodicalId":294701,"journal":{"name":"2015 IEEE Intelligent Vehicles Symposium (IV)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127865101","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 : 2015-08-27DOI: 10.1109/IVS.2015.7225786
L. Horne, J. Álvarez, M. Salzmann, N. Barnes
Efficient, fully-connected CRF inference enables fast semantic labelling of images. However, this requires high-quality unary potentials to be computed, which is currently time-consuming. While some recent work attempts to address this issue by only computing a subset of unary potentials, a need remains for a simple, fast way to decide which unary potentials should be computed, without sacrificing accuracy. In particular, for embedded applications, a method which avoids time or memory-intensive operations is desired. In this paper, we introduce an approach to selecting good locations to compute unary potentials. We implement an efficient morphological approach to select a small proportion of pixel locations where unary potentials will be calculated. The speed of our labelling method allows us to directly search a large parameter space to optimize our method for a given task. We show that our method can achieve comparable accuracy to what can be achieved when all unary potentials are calculated, with significant time saving. Furthermore, we show that it is possible to tune our method to yield improved accuracy for certain classes of interest. We demonstrate this over multiple datasets representing challenging applications for our approach.
{"title":"Efficient scene parsing by sampling unary potentials in a fully-connected CRF","authors":"L. Horne, J. Álvarez, M. Salzmann, N. Barnes","doi":"10.1109/IVS.2015.7225786","DOIUrl":"https://doi.org/10.1109/IVS.2015.7225786","url":null,"abstract":"Efficient, fully-connected CRF inference enables fast semantic labelling of images. However, this requires high-quality unary potentials to be computed, which is currently time-consuming. While some recent work attempts to address this issue by only computing a subset of unary potentials, a need remains for a simple, fast way to decide which unary potentials should be computed, without sacrificing accuracy. In particular, for embedded applications, a method which avoids time or memory-intensive operations is desired. In this paper, we introduce an approach to selecting good locations to compute unary potentials. We implement an efficient morphological approach to select a small proportion of pixel locations where unary potentials will be calculated. The speed of our labelling method allows us to directly search a large parameter space to optimize our method for a given task. We show that our method can achieve comparable accuracy to what can be achieved when all unary potentials are calculated, with significant time saving. Furthermore, we show that it is possible to tune our method to yield improved accuracy for certain classes of interest. We demonstrate this over multiple datasets representing challenging applications for our approach.","PeriodicalId":294701,"journal":{"name":"2015 IEEE Intelligent Vehicles Symposium (IV)","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125397708","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 : 2015-08-27DOI: 10.1109/IVS.2015.7225807
Kei Tateiwa, K. Yamada
This paper presents a method for estimating a driver's awareness of the presence of a pedestrian while crossing or trying to cross a crosswalk located in the path of a right or left turn when the driver is trying to turn right or left at an intersection based on the behavior of the vehicle operated by the driver. The method is based on the idea that different driving behaviors occur in similar situations depending on whether the driver notices a pedestrian. The results of an evaluation performed using actual driving behavior data of vehicles driven on public roads are also reported.
{"title":"Estimating driver awareness of pedestrians in crosswalk in the path of right or left turns at an intersection from vehicle behavior","authors":"Kei Tateiwa, K. Yamada","doi":"10.1109/IVS.2015.7225807","DOIUrl":"https://doi.org/10.1109/IVS.2015.7225807","url":null,"abstract":"This paper presents a method for estimating a driver's awareness of the presence of a pedestrian while crossing or trying to cross a crosswalk located in the path of a right or left turn when the driver is trying to turn right or left at an intersection based on the behavior of the vehicle operated by the driver. The method is based on the idea that different driving behaviors occur in similar situations depending on whether the driver notices a pedestrian. The results of an evaluation performed using actual driving behavior data of vehicles driven on public roads are also reported.","PeriodicalId":294701,"journal":{"name":"2015 IEEE Intelligent Vehicles Symposium (IV)","volume":"43 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2015-08-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127259502","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：