A maximal figure-of-merit learning approach to maximizing mean average precision with deep neural network based classifiers

Li, Kehuang; Huang, Zhen; Cheng, Yiwei; Lee, Chin‐Hui

doi:10.1109/icassp.2014.6854454

Cited by 40 publications

(16 citation statements)

References 22 publications

(21 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition, compared with that of YOLOv3-53, the loss function value of YOLOv3-13 has more and bigger fluctuations. As performance evaluation indicators, Intersection Over Union (IoU) and mAP are commonly used in target detection [36]. The IoU is often used for edge frame overlap or positioning accuracy evaluation.…”

Section: Model Training and Evaluationmentioning

confidence: 99%

“…During the experiment, through accuracy evaluation for two-category target detection, the mean accuracy AP was calculated first of all, and then the mAP value was obtained. As performance evaluation indicators, Intersection Over Union (IoU) and mAP are commonly used in target detection [36]. The IoU is often used for edge frame overlap or positioning accuracy evaluation.…”

Section: Model Training and Evaluationmentioning

confidence: 99%

See 1 more Smart Citation

Real-time Concealed Object Detection from Passive Millimeter Wave Images Based on the YOLOv3 Algorithm

Pang

Liu

Chen

et al. 2020

Sensors

View full text Add to dashboard Cite

The detection of objects concealed under people’s clothing is a very challenging task, which has crucial applications for security. When testing the human body for metal contraband, the concealed targets are usually small in size and are required to be detected within a few seconds. Focusing on weapon detection, this paper proposes using a real-time detection method for detecting concealed metallic weapons on the human body applied to passive millimeter wave (PMMW) imagery based on the You Only Look Once (YOLO) algorithm, YOLOv3, and a small sample dataset. The experimental results from YOLOv3-13, YOLOv3-53, and Single Shot MultiBox Detector (SSD) algorithm, SSD-VGG16, are compared ultimately, using the same PMMW dataset. For the perspective of detection accuracy, detection speed, and computation resource, it shows that the YOLOv3-53 model had a detection speed of 36 frames per second (FPS) and a mean average precision (mAP) of 95% on a GPU-1080Ti computer, more effective and feasible for the real-time detection of weapon contraband on human body for PMMW images, even with small sample data.

show abstract

Section: Model Training and Evaluationmentioning

confidence: 99%

Section: Model Training and Evaluationmentioning

confidence: 99%

Real-time Concealed Object Detection from Passive Millimeter Wave Images Based on the YOLOv3 Algorithm

Pang

Liu

Chen

et al. 2020

Sensors

View full text Add to dashboard Cite

show abstract

“…It was shown that discriminative learning outperforms a binary classification manner in automatic speech recognition and applied in minimum error classification [28] and minimum verification error [29]. The second approach explores the maximal figure-of-merit (MFoM) [30], [31] learning solution, which allows us to approximate the metrics of interest, namely the micro-F1 and equal error rate (EER), with a differentiable function, so that gradient-based optimization algorithms can be applied to learn DNN parameters. Specifically, MFoM tries to improve the decision boundary [30] using the output sigmoid scores without the need of any intermediate calibration.…”

Section: Motivationmentioning

confidence: 99%

Maximal Figure-of-Merit Framework to Detect Multi-Label Phonetic Features for Spoken Language Recognition

Kukanov

Trọng

Hautamäki

et al. 2020

IEEE/ACM Trans. Audio Speech Lang. Process.

View full text Add to dashboard Cite

Bottleneck features (BNFs) generated with a deep neural network (DNN) have proven to boost spoken language recognition accuracy over basic spectral features significantly. However, BNFs are commonly extracted using language-dependent tied-context phone states as learning targets. Moreover, BNFs are less phonetically expressive than the output layer in a DNN, which is usually not used as a speech feature because of its very high dimensionality hindering further post-processing. In this article, we put forth a novel deep learning framework to overcome all of the above issues and evaluate it on the 2017 NIST Language Recognition Evaluation (LRE) challenge. We use manner and place of articulation as speech attributes, which lead to low-dimensional "universal" phonetic features that can be defined across all spoken languages. To model the asynchronous nature of the speech attributes while capturing their intrinsic relationships in a given speech segment, we introduce a new training scheme for deep architectures based on a Maximal Figure of Merit (MFoM) objective. MFoM introduces non-differentiable metrics into the backpropagation-based approach, which is elegantly solved in the proposed framework. The experimental evidence collected on the recent NIST LRE 2017 challenge demonstrates the effectiveness of our solution. In fact, the performance of speech language recognition (SLR) systems based on spectral features is improved for more than 5% absolute Cavg. Finally, the F1 metric can be brought from 77.6% up to 78.1% by combining the conventional baseline phonetic BNFs with the proposed articulatory attribute features.

show abstract

“…where R denotes the number of images in a testing set, and Ra denotes the number of images that are correctly classified. Mean average precision (mAP) [33] is also used to measure the effectiveness of our method in multiclass classification.…”

Section: Weakly Supervised Localizationmentioning

confidence: 99%

Bilinear CNN Model for Fine-Grained Classification Based on Subcategory-Similarity Measurement

et al. 2019

View full text Add to dashboard Cite

One of the challenges in fine-grained classification is that subcategories with significant similarity are hard to be distinguished due to the equal treatment of all subcategories in existing algorithms. In order to solve this problem, a fine-grained image classification method by combining a bilinear convolutional neural network (B-CNN) and the measurement of subcategory similarities is proposed. Firstly, an improved weakly supervised localization method is designed to obtain the bounding box of the main object, which allows the model to eliminate the influence of background noise and obtain more accurate features. Then, sample features in the training set are computed by B-CNN so that the fuzzing similarity matrix for measuring interclass similarities can be obtained. To further improve classification accuracy, the loss function is designed by weighting triplet loss and softmax loss. Extensive experiments implemented on two benchmarks datasets, Stanford Cars-196 and Caltech-UCSD Birds-200-2011 (CUB-200-2011), show that the newly proposed method outperforms in accuracy several state-of-the-art weakly supervised classification models.

show abstract

A maximal figure-of-merit learning approach to maximizing mean average precision with deep neural network based classifiers

Cited by 40 publications

References 22 publications

Real-time Concealed Object Detection from Passive Millimeter Wave Images Based on the YOLOv3 Algorithm

Real-time Concealed Object Detection from Passive Millimeter Wave Images Based on the YOLOv3 Algorithm

Maximal Figure-of-Merit Framework to Detect Multi-Label Phonetic Features for Spoken Language Recognition

Bilinear CNN Model for Fine-Grained Classification Based on Subcategory-Similarity Measurement

Contact Info

Product

Resources

About