Gaussian YOLOv3: An Accurate and Fast Object Detector Using Localization Uncertainty for Autonomous Driving

Choi, Jung-Oh; Chun, Dayoung; Kim, Hyun; Lee, Hyuk-Jae

doi:10.1109/iccv.2019.00059

Cited by 424 publications

(222 citation statements)

References 25 publications

Supporting

Mentioning

221

Contrasting

Unclassified

Order By: Relevance

“…The above experiments show that even though this paper focuses mainly on YOLO detectors which achieve the best trade-offs between accuracy and speed [6], the proposed scheme also works well for various networks (i.e., YOLOv2, YOLOv3, ResNet50) on various datasets, including both object detection (i.e., VOC, COCO) and classification (i.e., ImageNet). Therefore, it is expected that the proposed scheme can be generalized to all other deep networks based on the convolutional layers.…”

Section: A Mixed Precision Quantizationmentioning

confidence: 94%

See 1 more Smart Citation

Layer-Specific Optimization for Mixed Data Flow With Mixed Precision in FPGA Design for CNN-Based Object Detectors

Nguyen

Kim

Lee

2021

IEEE Trans. Circuits Syst. Video Technol.

Self Cite

View full text Add to dashboard Cite

Convolutional neural networks (CNNs) require both intensive computation and frequent memory access, which lead to a low processing speed and large power dissipation. Although the characteristics of the different layers in a CNN are frequently quite different, previous hardware designs have employed common optimization schemes for them. This paper proposes a layer-specific design that employs different organizations that are optimized for the different layers. The proposed design employs two layer-specific optimizations: layer-specific mixed data flow and layer-specific mixed precision. The mixed data flow aims to minimize the off-chip access while demanding a minimal on-chip memory (BRAM) resource of an FPGA device. The mixed precision quantization is to achieve both a lossless accuracy and an aggressive model compression, thereby further reducing the offchip access. A Bayesian optimization approach is used to select the best sparsity for each layer, achieving the best trade-off between the accuracy and compression. This mixing scheme allows the entire network model to be stored in BRAMs of the FPGA to aggressively reduce the off-chip access, and thereby achieves a significant performance enhancement. The model size is reduced by 22.66-28.93 times compared to that in a full-precision network with a negligible degradation of accuracy on VOC, COCO, and ImageNet datasets. Furthermore, the combination of mixed dataflow and mixed precision significantly outperforms the previous works in terms of both throughput, off-chip access, and on-chip memory requirement.

show abstract

Section: A Mixed Precision Quantizationmentioning

confidence: 94%

“…It is noteworthy that Ti(l+1) = To(l) and Ml = Nl+1. Hence, Ti and To of each layer can be easily chosen by the guideline in (6).…”

Section: =1mentioning

confidence: 99%

Layer-Specific Optimization for Mixed Data Flow With Mixed Precision in FPGA Design for CNN-Based Object Detectors

Nguyen

Kim

Lee

2021

IEEE Trans. Circuits Syst. Video Technol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Combining other good ideas, YOLOv3 [19] improved prediction accuracy, especially for small objects, while maintaining its speed advantage. Based on the network of YOLOv3, Gaussian YOLOv3 [20] improved the performance of the model by increasing the output of the network and improving the loss function of the network.…”

Section: ) One-stage Methodsmentioning

confidence: 99%

Efficient Foreign Object Detection Between PSDs and Metro Doors via Deep Neural Networks

Dai

Liu

et al. 2020

IEEE Access

View full text Add to dashboard Cite

Platform Screen Doors (PSDs) have been widely used in modern Asian and European metro systems due to the advantages of safety, comfort for passengers. Unfortunately, someone or something will be caught by PSDs and metro doors occasionally, which may lead to serious accidents. Therefore, the foreign object detection between PSDs and metro doors is a burning problem. Moreover, this problem is a challenging and still largely under-explored topic. In recent years, we have seen significant improvements in generic object detection built on deep learning techniques. Accordingly, this paper adopts deep learning technologies to address the problem of foreign object detection between PSDs and metro doors. To the best of our knowledge, this is the first attempt to use deep learning to solve the problem. To realize this, a dataset including 984 real-world images (with 600 × 480 pixels) labeled for six types of foreign objects (bag, bottle, person, plastic bag, umbrella, other) is developed. Then, we compared the performance of some state-of-the-art object detection algorithms (such as You Only Look Once-YOLOv3, Single Shot MultiBox Detector-SSD, and CenterNet) on the dataset. Experimental results demonstrate that the foreign object detection algorithms based on deep neural networks have achieved excellent results, which not only improves the accuracy of detection but also give the categories of foreign objects. YOLOv3-tiny can achieve the fastest detection speed, up to 200 Frame Per Second (FPS); CenterNet can achieve the best detection results, up to 99.7% mean Average Precision (mAP).

show abstract

“…Most object detection networks could be divided into two categories: 1. two-stage deep neural networks, represented by R-CNN [31], including R-CNN, Fast R-CNN [32], Faster R-CNN [33], Mask R-CNN [34]; 2. one-stage deep neural networks with end-to-end structure, represented by YOLO [35] and its variants, including YOLO9000 [36], YOLOv3, Gaussian YOLOv3 [37], etc. Two-stage networks perform better, while one-stage networks operate faster.…”

Section: B Deep Learning-based Methodsmentioning

confidence: 99%

An Accurate Object Detector With Effective Feature Extraction by Intrinsic Prior Knowledge

Li¹,

Fang²,

Lu³

2020

IEEE Access

View full text Add to dashboard Cite

Deep learning-based object detection algorithms play an increasingly important role in many computer vision tasks. Many detection algorithms are trying to increase the depth of networks to improve their feature expression capacity. However, how to better extract the essential features from a fixed number of training samples is often ignored. Thus, we propose CoReSh network, using the prior knowledge of intrinsic features, to address this problem. First, the relevant optical factors which may affect the detection performance are explored, and the intrinsic features are proven to be able to represent the essential information of objects. Second, a new data augmentation is proposed to increase the feature diversities of training samples through intrinsic decomposition. Third, a deep detection neural network incorporating color, reflectance, and shading image's information is designed. To optimize the performance of our network, different learning rate setting tactics are compared. Besides, a contrast experiment on the Pascal VOC dataset was set up to verify it. The result shows that our network is 10% better than ordinary deep learning-based detection networks in precision, and roughly the same as the network loaded with pretraining weights from ILSVRC. It shows that our network has greatly improved the precision of object detection and it is suitable when the training process lacks sufficient training samples.

show abstract

Gaussian YOLOv3: An Accurate and Fast Object Detector Using Localization Uncertainty for Autonomous Driving

Cited by 424 publications

References 25 publications

Layer-Specific Optimization for Mixed Data Flow With Mixed Precision in FPGA Design for CNN-Based Object Detectors

Layer-Specific Optimization for Mixed Data Flow With Mixed Precision in FPGA Design for CNN-Based Object Detectors

Efficient Foreign Object Detection Between PSDs and Metro Doors via Deep Neural Networks

An Accurate Object Detector With Effective Feature Extraction by Intrinsic Prior Knowledge

Contact Info

Product

Resources

About