Detection of Traffic Signs Based on Combination of GAN and Faster-RCNN

Huang, Wenqing; Huang, Mingzhu; Zhang, Yuting

doi:10.1088/1742-6596/1069/1/012159

Cited by 14 publications

(11 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Pedestrian Detection [16] GAN for synthetic data creation [17] DCGAN + SSD [18] DCGAN + SSD [19] DCGAN + SSD Small Object Detection [20] Faster R-CNN + GAN [21] GAN + CNN + SSD/Faster R-CNN [22] CNN + ResNet + GAN Unsupervised Bounding Box Detection [23] CNN + GAN + Reinforcement Learning [24] Dilated CNN + GAN with Mask Mean Loss [25] Encoder + Conditional GAN 2.56 SSD300 Pascal VOC 2007 [16] Not Applicable Not Applicable Not Applicable [17] 45.2 SSD CIFAR-10/100 [18] 39.4 SSD VOC [19] Not Applicable Not Applicable Not Applicable [20] 19.47 Faster R-CNN Tsinghua-Tencent 100K [21] 25.1 FRCNN COWC Dataset [22] 60 Faster R-CNN (Small Objects) Tsinghua-Tencent 100K [23] Not Applicable Not Applicable Not Applicable [24] 5.37 [23] Car (Stanford) [25] 2.6 WCCN VGG16 VOC2007…”

Section: International Symposium On Innovation and Technology (Siintec)mentioning

confidence: 99%

“…Small Object Detection appears as an exponent application for GAN's. In [20] a GAN is embedded into a Faster R-CNN Network to generate residual representations of small objects to be similar to the ones of big objects which improves the detection ability of small objects when compared to a vanilla Faster R-CNN network. Results show around 19.5% better performance in small objects detection than a regular Faster R-CNN.…”

Section: International Symposium On Innovation and Technology (Siintec)mentioning

confidence: 99%

“…It then uses a different CNN to detect edges and improve the resolution even further to finally use SSD or a Faster R-CNN to detect the objects, improving the mAP performance in 25.1%. [22] approaches the small object detection task in a similar manner than [20], by using region proposals. The authors presented a perceptual GAN architecture where the generator creates super-resolved representations of small objects being supervised by the discriminator.…”

Section: International Symposium On Innovation and Technology (Siintec)mentioning

confidence: 99%

See 2 more Smart Citations

2d Image Object Detection Aided by Generative Adversarial Networks a Literature Review

Bertolini¹,

Monteiro²

2021

Blucher Engineering Proceedings

View full text Add to dashboard Cite

Object Detection (OD) is one of the most important tasks in 2D image processing. Multiple math models have been proposed and frameworks based in Deep Convolutional Networks such as R-CNN, SSD and YOLO are most common. Generative Adversarial Nets (GAN's) represent a prominent field of study in machine learning and it has been applied to many tasks with exciting results. The objective of this work is to assess the potential of GAN's applied to OD tasks and the proposed frameworks as field of study. The methodology used was a systemically review of 14 papers. The conclusion shows that even though OD and GAN's are popular themes, there are not many developments done in the intersection of both subjects. Therefore, OD with GAN applied tasks are an excellent field to explore in future works.

show abstract

Section: International Symposium On Innovation and Technology (Siintec)mentioning

confidence: 99%

Section: International Symposium On Innovation and Technology (Siintec)mentioning

confidence: 99%

Section: International Symposium On Innovation and Technology (Siintec)mentioning

confidence: 99%

See 1 more Smart Citation

2d Image Object Detection Aided by Generative Adversarial Networks a Literature Review

Bertolini¹,

Monteiro²

2021

Blucher Engineering Proceedings

View full text Add to dashboard Cite

show abstract

“…References [16], [17], [37] utilize the context information surrounding objects to increase classification accuracy. References [38], [39] achieve better detection of small objects by using GAN to generate super-resolved representations for them. Reference [17] extracts local features to achieve fine-grained classification.…”

Section: Traffic Sign Detectionmentioning

confidence: 99%

Traffic Sign Detection and Recognition Using Novel Center-Point Estimation and Local Features

Wei

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Traffic sign detection is one of the critical technologies in the field of intelligent transportation systems (ITS). The difficulty of traffic sign detection mainly lies in detecting small objects in a wide and complex traffic scene quickly and accurately. In this paper, we regard traffic sign detection as a region classification problem and propose a two-stage CNN-based approach to solve it. At the first stage, we design an efficient network which is built with improved fire-modules to generate object proposals quickly. The network up-samples and merges the feature maps of different scales to attain a high-resolution fused feature map which contains semantically strong features of multi-scale objects. Specially, the prediction is made on the fuse feature map and based on the novel center-point estimation. With the overall designs, our region proposal network can achieve high recall value while using low-resolution images. At the second stage, a separate classification network is proposed. The bottleneck of the classification performance is generally caused by the greatly similar appearances between traffic signs. Therefore, we further explore local regions with critical differences between traffic signs to obtain fine-grained local features which help to improve classification. Finally, we evaluate our method on a challenge benchmark Tsinghua-Tencent 100K which provides many large images with small traffic sign instances. The experiment result shows that our method has better performance and faster detection speed than many state-of-the-art traffic sign detection methods. INDEX TERMS Traffic sign detection, multi-scale, center-point estimation, local features.

show abstract

“…Liu et al [17] and Liang et al [41] used different levels of features to improve detection of small traffic signs. Huang et al [42] and Li et al [16] generated features of small objects with the same representation capability as those from large objects through GAN. Huang et al [43] pointed out that fine-grained classification is a key problem in traffic sign detection, and proposed a two-stage detection scheme to improve traffic sign detection.…”

Section: B Traffic Sign Detectionmentioning

confidence: 99%

FAMN: Feature Aggregation Multipath Network for Small Traffic Sign Detection

et al. 2019

View full text Add to dashboard Cite

Traffic sign detection has achieved promising results in recent years. Nevertheless, there are still two problems remain to be overcome. One problem is the detection of small traffic signs, which usually occupy less than 2% of the image area. The other problem is fine-grained classification, with difficulties arising from similar appearances between traffic signs. For example, different speed-limit traffic signs have differences solely from the speed numbers. In this paper, we propose a Feature Aggregation MultiPath Network (FAMN) to tackle the problems simultaneously. First, we propose a Feature Aggregation (FA) structure to aggregate regional features from different feature maps by using element-wise Max, then convolution layers are used to extract rich semantic features. Accordingly, objects of different scales can choose the best features to improve performance of small object detection. Second, we propose a Multipath Network (MN) structure to obtain fine-grained features. The MN structure consists of three paths, extracting instance-level, part-level, and context-level features, respectively. The three types of features are then concatenated to form fine-grained features of the proposals. Experimental results demonstrate the effectiveness of our proposed FAMN. Specifically, FAMN is able to obtain an average F1-measure of 93.1% in TT100K dataset, 2.9% higher than the state-of-the-art.

show abstract

Detection of Traffic Signs Based on Combination of GAN and Faster-RCNN

Cited by 14 publications

References 1 publication

2d Image Object Detection Aided by Generative Adversarial Networks a Literature Review

2d Image Object Detection Aided by Generative Adversarial Networks a Literature Review

Traffic Sign Detection and Recognition Using Novel Center-Point Estimation and Local Features

FAMN: Feature Aggregation Multipath Network for Small Traffic Sign Detection

Contact Info

Product

Resources

About