Beyond Trade-Off: Accelerate FCN-Based Face Detector with Higher Accuracy

Song, Guanglu; Liu, Yu; Jiang, Ming; Wang, Yujie; Yan, Junjie; Leng, Biao

doi:10.1109/cvpr.2018.00809

Cited by 40 publications

(28 citation statements)

References 39 publications

(59 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4 shows the specific structure of the Masked RPN. On the basis of the original RPN, each convolution layer is replaced by the masked convolution layer [28], [29]. In masked convolution, only the region of interest of the image is convoluted, and other regions are simply set to zero, thereby reducing part of the matrix point multiplication.…”

Section: B Color-guided Anchoring Strategymentioning

confidence: 99%

Information-Guided Flame Detection Based on Faster R-CNN

2020

View full text Add to dashboard Cite

Due to the diversity of the shape and texture of flame, and interference objects that similar to flame in color, detecting the position of flame from images is a difficult task. To enable generic object detection methods to achieve better performance in flame detection tasks, a color-guided anchoring strategy is proposed that uses color features of the flame to limit the location of the anchor. To solve the problem of high false alarm rate when directly using generic object detection methods in flame detection, a global information-guided flame detection method is proposed, this strategy uses a parallel network to generate image global information. We use these two methods to improve Faster R-CNN (Regions with Convolutional Neural Network features) to perform the fire detection process in a guided manner. Experiments on the BoWFire dataset show that our method improved detection speed by 10.1% compared with the original Faster R-CNN. In addition, the false alarm rate is decreased by 21.5%, and the overall accuracy of detection is increased by 9.3%. Experiments on PascalVOC and Corsician datasets further demonstrate the robustness of the proposed methods.

show abstract

Section: B Color-guided Anchoring Strategymentioning

confidence: 99%

Information-Guided Flame Detection Based on Faster R-CNN

2020

View full text Add to dashboard Cite

show abstract

“…The specific evaluation metrics include accuracy, error, and RMSE are utilized to compare the performance of various methods in the experiments. Compared to [45], our evaluation metric is more comprehensive and the performance is very impressive. Table I shows the overall results of ablation experiment.…”

Section: B Ablation and Comparison Experimentsmentioning

confidence: 99%

DFENet: Deep Feature Enhancement Network for Accurate Calculation of Instantaneous Wave-Free Ratio

Song

Zhang

2020

IEEE J. Transl. Eng. Health Med.

View full text Add to dashboard Cite

Accurate iFR calculation can provide important clinical information for intracoronary functional assessment without administration of adenosine, which needs to locate object points in the pressure waveforms: peak, the dichrotic notch and the pressure nadir at the end of diastole. We propose a DFENet that is capable of locating object points to calculate iFR accurately. We first design a SFRA into DFENet with the idea of DenseNet. To avoid overfitting when dealing with sparse signals, we set appropriate number of network layers, growth rate of dense blocks and compression rate of transition blocks in 1D DenseNet. Then, we introduce a feature enhancement mechanism named 1D SE block for enhancing inconspicuous but vital features from SFRA, which guides DFENet to focus on these important features via feature recalibration. Finally, we prove an effective interaction mode between SFRA and 1D SE block to locate object points accurately. Adequate experiments demonstrate that DFENet reaches a high accuracy of 94.22%, error of 5.6 on object point localization of 1D pressure waveforms that include 1457 samples from 100 subjects via a cross-validation of Leave-One-Out. Comparison experiment demonstrates that the accuracy of DFENet exceeds other state-of-the-art methods by 3.35%, and ablation experiment demonstrates that the accuracy of SFRA and cSE exceed the other variations by 6.63% and 2.56% respectively. Importantly, we reveal how the DFENet enhance inconspicuous but vital feature by applying gradient-weighted class activation maps. DFENet can locate object points accurately, which is applicable to other signal processing tasks, especially in health sensing.

show abstract

“…Table 2 compares the speed and performance of fast detectors. For the most popular PASCAL VOC dataset, amongst the Yolo models, Fast YOLO proves to be the fastest [21]. But YOLO is more precise when compared to the faster version of YOLO.…”

Section: Frameworkmentioning

confidence: 99%

“…Fast R-CNN makes more background errors and fewer localization errors [4]. Table 3 is the comparison of YOLOv2 with other detection frameworks for PASCAL VOC 2007 [21]. The second version of YOLO is more accurate and fast when compared to the previous detection systems.…”

Section: Performancementioning

confidence: 99%

Evaluation and Evolution of Object Detection Techniques YOLO and R-CNN

2019

ijrte

View full text Add to dashboard Cite

Object detection has boomed in areas like image processing in accordance with the unparalleled development of CNN (Convolutional Neural Networks) over the last decade. The CNN family which includes R-CNN has advanced to much faster versions like Fast-RCNN which have mean average precision(Map) of up to 76.4 but their frames per second(fps) still remain between 5 to 18 and that is comparatively moderate to problem-solving time. Therefore, there is an urgent need to increase speed in the advancements of object detection. In accordance with the broad initiation of CNN and its features, this paper discusses YOLO (You only look once), a strong representative of CNN which comes up with an entirely different method of interpreting the task of detecting the objects. YOLO has attained fast speeds with fps of 155 and map of about 78.6, thereby surpassing the performances of other CNN versions appreciably. Furthermore, in comparison with the latest advancements, YOLOv2 attains an outstanding trade-off between accuracy and speed and also as a detector possessing powerful generalization capabilities of representing an entire image

show abstract

Beyond Trade-Off: Accelerate FCN-Based Face Detector with Higher Accuracy

Cited by 40 publications

References 39 publications

Information-Guided Flame Detection Based on Faster R-CNN

Information-Guided Flame Detection Based on Faster R-CNN

DFENet: Deep Feature Enhancement Network for Accurate Calculation of Instantaneous Wave-Free Ratio

Evaluation and Evolution of Object Detection Techniques YOLO and R-CNN

Contact Info

Product

Resources

About