SqueezeJet: High-Level Synthesis Accelerator Design for Deep Convolutional Neural Networks

A large number of videos are uploaded on educational platforms every minute. Those platforms are responsible for any sensitive media uploaded by their users. An automated detection system to identify pornographic content could assist human workers by pre-selecting suspicious videos. In this paper, we propose a multimodal approach to adult content detection. We use two Deep Convolutional Neural Networks to extract high-level features from both image and audio sources of a video. Then, we concatenate those features and evaluate the performance of classifiers on a set of mixed educational and pornographic videos. We achieve an F1-score of 95.67% on the educational and adult videos set and an F1-score of 94% on our test subset for the pornographic class.

Download Full-text

Meaning maps and saliency models based on deep convolutional neural networks are insensitive to image meaning when predicting human fixations

10.1101/840256 ◽

2019 ◽

Author(s):

Marek A. Pedziwiatr ◽

Matthias Kümmerer ◽

Thomas S.A. Wallis ◽

Matthias Bethge ◽

Christoph Teufel

Keyword(s):

Neural Network ◽

Neural Networks ◽

Eye Movements ◽

Convolutional Neural Networks ◽

Deep Neural Network ◽

Image Features ◽

Human Vision ◽

Deep Convolutional Neural Networks ◽

High Level

AbstractEye movements are vital for human vision, and it is therefore important to understand how observers decide where to look. Meaning maps (MMs), a technique to capture the distribution of semantic importance across an image, have recently been proposed to support the hypothesis that meaning rather than image features guide human gaze. MMs have the potential to be an important tool far beyond eye-movements research. Here, we examine central assumptions underlying MMs. First, we compared the performance of MMs in predicting fixations to saliency models, showing that DeepGaze II – a deep neural network trained to predict fixations based on high-level features rather than meaning – outperforms MMs. Second, we show that whereas human observers respond to changes in meaning induced by manipulating object-context relationships, MMs and DeepGaze II do not. Together, these findings challenge central assumptions underlying the use of MMs to measure the distribution of meaning in images.

Download Full-text

Learning to Automatically Diagnose Multiple Diseases in Pediatric Chest Radiographs Using Deep Convolutional Neural Networks

10.1101/2021.08.12.21261954 ◽

2021 ◽

Author(s):

Thanh T. Tran ◽

Hieu H. Pham ◽

Thang V. Nguyen ◽

Tung T. Le ◽

Hieu T. Nguyen ◽

...

Keyword(s):

Neural Networks ◽

Convolutional Neural Networks ◽

Class Imbalance ◽

Learning Task ◽

Deep Convolutional Neural Networks ◽

Thoracic Diseases ◽

Experienced Radiologist ◽

Pediatric Chest ◽

High Level ◽

Sensitivity Specificity

Chest radiograph (CXR) interpretation is critical for the diagnosis of various thoracic diseases in pediatric patients. This task, however, is error-prone and requires a high level of understanding of radiologic expertise. Recently, deep convolutional neural networks (D-CNNs) have shown remarkable performance in interpreting CXR in adults. However, there is a lack of evidence indicating that D-CNNs can recognize accurately multiple lung pathologies from pediatric CXR scans. In particular, the development of diagnostic models for the detection of pediatric chest diseases faces significant challenges such as (i) lack of physician-annotated datasets and (ii) class imbalance problems. In this paper, we retrospectively collect a large dataset of 5,017 pediatric CXR scans, for which each is manually labeled by an experienced radiologist for the presence of 10 common pathologies. A D-CNN model is then trained on 3,550 annotated scans to classify multiple pediatric lung pathologies automatically. To address the high-class imbalance issue, we propose to modify and apply "Distribution-Balanced loss" for training D-CNNs which reshapes the standard Binary-Cross Entropy loss (BCE) to efficiently learn harder samples by down-weighting the loss assigned to the majority classes. On an independent test set of 777 studies, the proposed approach yields an area under the receiver operating characteristic (AUC) of 0.709 (95% CI, 0.690-0.729). The sensitivity, specificity, and F1-score at the cutoff value are 0.722 (0.694-0.750), 0.579 (0.563-0.595), and 0.389 (0.373-0.405), respectively. These results significantly outperform previous state-of-the-art methods on most of the target diseases. Moreover, our ablation studies validate the effectiveness of the proposed loss function compared to other standard losses, e.g., BCE and Focal Loss, for this learning task. Overall, we demonstrate the potential of D-CNNs in interpreting pediatric CXRs.

Download Full-text