Traffic Scene Classification on a Representation Budget

Sikirić, Ivan; Brkić, Karla; Bevandić, Petra; Krešo, Ivan; Krapac, Josip; Šegvić, Siniša

doi:10.1109/tits.2019.2891995

Cited by 14 publications

(10 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Honda [7] and FM3 [10] datasets focus on driving scene categorization, and Honda defines 12 typical traffic scenarios: branch left, branch right, construction zone, merge left, 3&4&4-way intersection, overhead bridge, rail crossing, tunnel, zebra crossing, where FM3 defines 8 traffic scene: highway, road, tunnel, exit, settlement, overpass, booth, traffic. They are both collected by an on-board RGB camera and annotated by human operators.…”

Section: B Scene Category Definitionmentioning

confidence: 99%

“…MIT place [8] and Large Scene understanding (LSUN) [9] datasets contain a large number of labeled images that were originally downloaded from websites by using Google and other image search engines. Honda [7] and FM3 [10] datasets are more motivated by robotic and autonomous driving applications, which contain egocentric videos of driving scenes that are labeled at the frame level. Scene labels (i.e.…”

Section: Introductionmentioning

confidence: 99%

“…cafeteria, train station platform [8], intersection, construction zone [7]; (2) a target object of the scene, e.g. people, bus [9], overhead bridge [10]; (3) the static status of the target object or scene, e.g. road surface, weather condition [7]; (4) the dynamic status of the agent with respect to the scene, e.g.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

An Image-based Approach of Task-driven Driving Scene Categorization

Hu,

Fan,

Gao

et al. 2021

Preprint

View full text Add to dashboard Cite

Categorizing driving scenes via visual perception is a key technology for safe driving and the downstream tasks of autonomous vehicles. Traditional methods infer scene category by detecting scene-related objects or using a classifier that is trained on large datasets of fine-labeled scene images. Whereas at cluttered dynamic scenes such as campus or park, human activities are not strongly confined by rules, and the functional attributes of places are not strongly correlated with objects. So how to define, model and infer scene categories is crucial to make the technique really helpful in assisting a robot to pass through the scene. This paper proposes a method of task-driven driving scene categorization using weakly supervised data. Given a front-view video of a driving scene, a set of anchor points is marked by following the decision making of a human driver, where an anchor point is not a semantic label but an indicator meaning the semantic attribute of the scene is different from that of the previous one. A measure is learned to discriminate the scenes of different semantic attributes via contrastive learning, and a driving scene profiling and categorization method is developed based on that measure. Experiments are conducted on a front-view video that is recorded when a vehicle passed through the cluttered dynamic campus of Peking University. The scenes are categorized into straight road, turn road and alerting traffic. The results of semantic scene similarity learning and driving scene categorization are extensively studied, and positive result of scene categorization is 97.17 % on the learning video and 85.44% on the video of new scenes.

show abstract

Section: B Scene Category Definitionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

An Image-based Approach of Task-driven Driving Scene Categorization

Hu,

Fan,

Gao

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Recently, with increasing research interests in traffic scene understanding, some large-scale datasets [19], [27]- [30] have been published in the past few years. Besides, several methods [19], [30]- [33] focus on the deep neural architectures, inspired by the great success of image classification. [33] developed a multi-label classification framework that is suitable for many mobile agents through short image descriptors.…”

Section: A Scene Recognitionmentioning

confidence: 99%

Context-Aware Multi-Task Learning for Traffic Scene Recognition in Autonomous Vehicles

Lee

Jeon

et al. 2020

Preprint

View full text Add to dashboard Cite

Traffic scene recognition, which requires various visual classification tasks, is a critical ingredient in autonomous vehicles. However, most existing approaches treat each relevant task independently from one another, never considering the entire system as a whole. Because of this, they are limited to utilizing a task-specific set of features for all possible tasks of inference-time, which ignores the capability to leverage common task-invariant contextual knowledge for the task at hand. To address this problem, we propose an algorithm to jointly learn the task-specific and shared representations by adopting a multi-task learning network. Specifically, we present a lower bound for the mutual information constraint between shared feature embedding and input that is considered to be able to extract common contextual information across tasks while preserving essential information of each task jointly. The learned representations capture richer contextual information without additional task-specific network. Extensive experiments on the large-scale dataset HSD demonstrate the effectiveness and superiority of our network over state-of-the-art methods.

show abstract

“…In the field of image classification, many scholars have proposed their methods. Pang et al proposed a novel fused CNN that fused the features extracted in the shallow and deep layers for the biomedical image classification [9]; Ivan et al studied an image classification method on a tight representation budget, it focused on very short image descriptor which might be lost during the training process [10]; Zhou et al proposed a novel data augmentation strategy for the Siamese model and introduced a joint decision mechanism into the model which can better improve the classification performance [11]. All the above methods are well-behaved in the image classification filed and widely applied in many different fields.…”

Section: Introductionmentioning

confidence: 99%

A Dilated CNN Model for Image Classification

2019

View full text Add to dashboard Cite

The dilated convolution algorithm, which is widely used for image segmentation, is applied in the image classification field in this paper. In many traditional image classification algorithms, convolution neural network (CNN) plays an important role. However, the classical CNN has the problem of consuming too much computing resources. To solve this problem, first, this paper proposed a dilated CNN model which is built through replacing the convolution kernels of traditional CNN by the dilated convolution kernels, and then, the dilated CNN model is tested on the Mnist handwritten digital recognition data set. Second, to solve the detail loss problem in the dilated CNN model, the hybrid dilated CNN (HDC) is built by stacking dilated convolution kernels with different dilation rates, and then the HDC model is tested on the wide-band remote sensing image data set of earth's terrain. The results show that under the same environment, compared with the traditional CNN model, the dilated CNN model reduces the training time by 12.99% and improves the training accuracy by 2.86% averagely, compared with the dilated CNN model, the HDC model reduces the training time by 2.02% and improves the training and testing accuracy by 14.15% and 15.35% averagely. Therefore, the dilated CNN and HDC model proposed in this paper can significantly improve the image classification performance. INDEX TERMS Image classification, CNN, dilated convolution, hybrid dilated CNN.

show abstract

Traffic Scene Classification on a Representation Budget

Cited by 14 publications

References 28 publications

An Image-based Approach of Task-driven Driving Scene Categorization

An Image-based Approach of Task-driven Driving Scene Categorization

Context-Aware Multi-Task Learning for Traffic Scene Recognition in Autonomous Vehicles

A Dilated CNN Model for Image Classification

Contact Info

Product

Resources

About