IceVisionSet: lossless video dataset collected on Russian winter roads with traffic sign annotations

Pavlov, Artem L.; Karpyshev, Pavel; Ovchinnikov, George; Oseledets, Ivan; Tsetserukou, Dzmitry

doi:10.1109/icra.2019.8794341

Cited by 15 publications

(12 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Unlike most datasets, such as ImageNet and COCO, which collect pictures through online keyword indexing, or Tsinghua-Tencent 100 K, which collects pictures through Tencent Street View, the original pictures of our dataset were selected from the traffic photos marked IceVisonSet [15]. The picture of the sign lights and storage method is from street views of the real snow and ice season in Russia.…”

Section: Data Collectionmentioning

confidence: 99%

“…According to the size of the signs, we divided the included traffic signs into small, medium, and large, as shown in Table 1. The ice and snow environment is not only just a representative environment, but also has a strong influence on the brightness, completeness, color saturation, and other information of the picture, which have been proved to have a great impact on traffic sign detection and classification [15,29]. The dataset includes these types of pictures and pictures of day and night, shown in Figure 1, which are more challenging for TSR problem not included in most public dataset like GTSRB.…”

Section: Itsdbmentioning

confidence: 99%

“…Problems, such as the size and occlusion of traffic signs in environments, makes it more difficult to obtain information from the pictures, and presents greater challenges in the detection and classification of traffic signs. A Russian team [15] published a Russian ice and snow environment automatic driving dataset called IceVisionSet containing incomplete traffic sign annotations; however, the research focused on the light effects on a traffic sign recognition model in ice and snow environments and did not propose a useful model.…”

Section: Introductionmentioning

confidence: 99%

“…At the same time, the traditional method of simultaneous detection and classification is often not effective in such cases. In this article, our research goal is to achieve a better classification accuracy in the ice and snow dataset based on [15] through our design. The contribution points of this article are as follows:…”

Section: Introductionmentioning

confidence: 99%

“…According to the traffic pictures collected under the ice and snow environment of [15], we produced a more challenging ice and snow environment traffic sign detection dataset (ice environment traffic sign detection benchmark (ITSDB)) and a snow and ice environment traffic sign classification dataset (ice environment traffic sign recognition benchmark (ITSRB)) that both contain different sizes and complex environmental factors:…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Learning Region-Based Attention Network for Traffic Sign Recognition

Zhou

Zhan

2021

Sensors

View full text Add to dashboard Cite

Traffic sign recognition in poor environments has always been a challenge in self-driving. Although a few works have achieved good results in the field of traffic sign recognition, there is currently a lack of traffic sign benchmarks containing many complex factors and a robust network. In this paper, we propose an ice environment traffic sign recognition benchmark (ITSRB) and detection benchmark (ITSDB), marked in the COCO2017 format. The benchmarks include 5806 images with 43,290 traffic sign instances with different climate, light, time, and occlusion conditions. Second, we tested the robustness of the Libra-RCNN and HRNetv2p on the ITSDB compared with Faster-RCNN. The Libra-RCNN performed well and proved that our ITSDB dataset did increase the challenge in this task. Third, we propose an attention network based on high-resolution traffic sign classification (PFANet), and conduct ablation research on the design parallel fusion attention module. Experiments show that our representation reached 93.57% accuracy in ITSRB, and performed as well as the newest and most effective networks in the German traffic sign recognition dataset (GTSRB).

show abstract

Section: Data Collectionmentioning

confidence: 99%

Section: Itsdbmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Learning Region-Based Attention Network for Traffic Sign Recognition

Zhou

Zhan

2021

Sensors

View full text Add to dashboard Cite

show abstract

Robotics in Snow and Ice

Pomerleau

2022

Encyclopedia of Robotics

View full text Add to dashboard Cite

Brno urban dataset: Winter extension

2022

View full text Add to dashboard Cite

This paper presents our latest extension of the Brno Urban Dataset (BUD), the Winter Extension (WE). The dataset contains data from commonly used sensors in the automotive industry, like four RGB and single IR cameras, three 3D LiDARs, differential RTK GNSS receiver with heading estimation, the IMU and FMCW radar. Data from all sensors are precisely timestamped for future offline interpretation and data fusion. The most significant gain of the dataset is the focus on the winter conditions in snow-covered environments. Only a few public datasets deal with these kinds of conditions. We recorded the dataset during February 2021 in Brno, Czechia, when fresh snow covers the entire city and the surrounding countryside. The dataset contains situations from the city center, suburbs, highways as well as the countryside. Overall, the new extension adds three hours of real-life traffic situations from the mid-size city to the existing 10 h of original records. Additionally, we provide the precalculated YOLO neural network object detection annotations for all five cameras for the entire old data and the new ones. The dataset is suitable for developing mapping and navigation algorithms as well as the collision and object detection pipelines. The entire dataset is available as open-source under the MIT license.

show abstract

IceVisionSet: lossless video dataset collected on Russian winter roads with traffic sign annotations

Cited by 15 publications

References 19 publications

Learning Region-Based Attention Network for Traffic Sign Recognition

Learning Region-Based Attention Network for Traffic Sign Recognition

Robotics in Snow and Ice

Brno urban dataset: Winter extension

Contact Info

Product

Resources

About