Detection of traffic signs in real-world images: The German traffic sign detection benchmark

By Sebastian Houben, Johannes Stallkamp, Jan Salmen, Marc Schlipsing, Christian Igel. 2013

1,075 citations recorded by the source corpus.

Subfields: Civil and Structural Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Civil and Structural Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Repurposing Pretrained Models for Robust Out-of-domain Few-Shot Learning (2021, methodology): “We evaluate our model on the test split of miniImageNet – for the samedomain setting – as well as CUB-200-2011 (Welinder et al., 2010), Traffic Sign (Houben et al., 2013) and VGG Flower (Nilsback & Zisserman, 2008) – for the cross-domain setting.”
Meta-Dataset: A Dataset of Datasets for Learning to Learn from Few Examples (2019, positive): “Moreover, two of these datasets, Trafﬁc Signs and MSCOCO, are fully reserved for evaluation, meaning that no classes from them participate in the training set.”
Deep Clustered Convolutional Kernels (2015, neutral): “Experimental results confirmed this process results in gradually improving performance, while the training algorithm jointly optimizes structure as well as model’s parameters.”
Object Detection in 20 Years: A Survey (2019, background): “Image from GTSRB Dataset [81].”
Invisible Backdoor Attacks on Deep Neural Networks Via Steganography and Regularization (2019, background): “4 Network 6CNN+2FC 6CNN+2FC 6CNN+2FC 6CNN+2FC Dataset GTSRB GTSRB GTSRB GTSRB Target classes [1-43] [1-43] [1-43] [1-43] Trigger Number 43 43 43 43 Inject Rate (Each Class) 0.”
Rogue Signs: Deceiving Traffic Sign Recognition with Malicious Ads and Logos (2018, methodology): “The entire recognition pipeline is tested with the German Trafﬁc Sign Detection Benchmark (GTSDB) [23] where it achieves a reasonable mean average precision (mAP) of 75.59% at the intersection-over-union (IoU) of 0.5.”
EMBER: An Open Dataset for Training Static PE Malware Machine Learning Models (2018, background): “These include handwritten digit classification (e.g., MNIST [17]), image labeling (e.g., CIFAR [16] or Ima-geNet [10]), traffic sign detection [13], speech recognition (e.g., TIMIT [32]), sentiment analysis (e.g., Sentiment140 [12]), and a host of other datasets suitable for training models to…”
CURE-TSR: Challenging Unreal and Real Environments for Traffic Sign Recognition (2017, background): “It is a common practice to test the robustness of these systems with traffic datasets [1, 2, 3, 4, 5, 6, 7, 8, 9, 10].”
Traffic sign recognition — How far are we from the solution? (2013, methodology): “To assess detection performance we use GTSD and BTSD.”
A New Backdoor Attack in CNNS by Training Set Corruption Without Label Poisoning (2019, methodology): “For the experiments of trafﬁc signs classiﬁcation, we considered 16 different classes (corresponding to the most populated classes) from the German Trafﬁc Sign (GTSRB) benchmark dataset [10]; the 16 classes consist of 6 speed limit signs, 3 prohibition signs, 3 danger signs and 4 mandatory signs.”
Improved Few-Shot Visual Classification (2019, neutral): “Meta-Dataset [39] is a benchmark for few-shot learning encompassing 10 labeled image datasets: ILSVRC2012 (ImageNet) [31], Omniglot [18], FGVC-Aircraft (Aircraft) [22], CUB-200-2011 (Birds) [41], Describable Textures (DTD) [2], QuickDraw [14], FGVCx Fungi (Fungi) [35], VGG Flower (Flower) [25], Traf”
Traffic sign detection for U.S. roads: Remaining challenges and a case for tracking (2014, background): “Recently, the same authors endeavored to repeat the success for detection with the GTSDB [2].”
Perceptual-Sensitive GAN for Generating Adversarial Patches (2019, methodology): “To evaluate the effectiveness of the proposed method, extensive experiments are conducted on GTSRB (Houben et al. 2008) and ImageNet (Deng et al.”
Optimized Generic Feature Learning for Few-shot Classification across Domains (2020, background): “It consists of 10 datasets from diverse domains: ILSVRC-2012 [45] (the ImageNet dataset, consisting of natural images with 1000 categories), Omniglot [28] (hand-written characters, 1623 classes), Aircraft [34] (dataset of aircraft images, 100 classes) , CUB200-2011 [58] (dataset of Birds, 200 classe”
Towards Large-Scale Small Object Detection: Survey and Benchmarks (2022, background): “GTSDB [150] Traffic sign detection IJCNN 2013 900 1206 A benchmark for traffic sign detection collected under different scenarios”