ERNet: An Efficient and Reliable Human-Object Interaction Detection Network.

Human-Object Interaction (HOI) detection recognizes how persons interact with objects, which is advantageous in autonomous systems such as self-driving vehicles and collaborative robots. However, current HOI detectors are often plagued by model inefficiency and unreliability when making a prediction, which consequently limits its potential for real-world scenarios. In this paper, we address these challenges by proposing ERNet, an end-to-end trainable convolutional-transformer network for HOI detection. The proposed model employs an efficient multi-scale deformable attention to effectively capture vital HOI features. We also put forward a novel detection attention module to adaptively generate semantically rich instance and interaction tokens. These tokens undergo pre-emptive detections to produce initial region and vector proposals that also serve as queries which enhances the feature refinement process in the transformer decoders. Several impactful enhancements are also applied to improve the HOI representation learning. Additionally, we utilize a predictive uncertainty estimation framework in the instance and interaction classification heads to quantify the uncertainty behind each prediction. By doing so, we can accurately and reliably predict HOIs even under challenging scenarios. Experiment results on the HICO-Det, V-COCO, and HOI-A datasets demonstrate that the proposed model achieves state-of-the-art performance in detection accuracy and training efficiency. Codes are publicly available at https://github.com/Monash-CyPhi-AI-Research-Lab/ernet.

Risk score prediction model based on single nucleotide polymorphism for predicting malaria: a machine learning approach.

BACKGROUND: The malaria risk prediction is currently limited to using advanced statistical methods, such as time series and cluster analysis on epidemiological data. Nevertheless, machine learning models have been explored to study the complexity of malaria through blood smear images and environmental data. However, to the best of our knowledge, no study analyses the contribution of Single Nucleotide Polymorphisms (SNPs) to malaria using a machine learning model. More specifically, this study aims to quantify an individual's susceptibility to the development of malaria by using risk scores obtained from the cumulative effects of SNPs, known as weighted genetic risk scores (wGRS). RESULTS: We proposed an SNP-based feature extraction algorithm that incorporates the susceptibility information of an individual to malaria to generate the feature set. However, it can become computationally expensive for a machine learning model to learn from many SNPs. Therefore, we reduced the feature set by employing the Logistic Regression and Recursive Feature Elimination (LR-RFE) method to select SNPs that improve the efficacy of our model. Next, we calculated the wGRS of the selected feature set, which is used as the model's target variables. Moreover, to compare the performance of the wGRS-only model, we calculated and evaluated the combination of wGRS with genotype frequency (wGRS + GF). Finally, Light Gradient Boosting Machine (LightGBM), eXtreme Gradient Boosting (XGBoost), and Ridge regression algorithms are utilized to establish the machine learning models for malaria risk prediction. CONCLUSIONS: Our proposed approach identified SNP rs334 as the most contributing feature with an importance score of 6.224 compared to the baseline, with an importance score of 1.1314. This is an important result as prior studies have proven that rs334 is a major genetic risk factor for malaria. The analysis and comparison of the three machine learning models demonstrated that LightGBM achieves the highest model performance with a Mean Absolute Error (MAE) score of 0.0373. Furthermore, based on wGRS + GF, all models performed significantly better than wGRS alone, in which LightGBM obtained the best performance (0.0033 MAE score).

Improved Coefficient Recovery and Its Application for Rewritable Data Embedding.

JPEG is the most commonly utilized image coding standard for storage and transmission purposes. It achieves a good rate-distortion trade-off, and it has been adopted by many, if not all, handheld devices. However, often information loss occurs due to transmission error or damage to the storage device. To address this problem, various coefficient recovery methods have been proposed in the past, including a divide-and-conquer approach to speed up the recovery process. However, the segmentation technique considered in the existing method operates with the assumption of a bi-modal distribution for the pixel values, but most images do not satisfy this condition. Therefore, in this work, an adaptive method was employed to perform more accurate segmentation, so that the real potential of the previous coefficient recovery methods can be unleashed. In addition, an improved rewritable adaptive data embedding method is also proposed that exploits the recoverability of coefficients. Discrete cosine transformation (DCT) patches and blocks for data hiding are judiciously selected based on the predetermined precision to control the embedding capacity and image distortion. Our results suggest that the adaptive coefficient recovery method is able to improve on the conventional method up to 27% in terms of CPU time, and it also achieved better image quality with most considered images. Furthermore, the proposed rewritable data embedding method is able to embed 20,146 bits into an image of dimensions 512×512.

CUPSEED - A combined use of prediction syntax elements to embed data in SHVC video.

With the rapid advancement in digital technologies, video rises to become one of the most effective communication tools that continues to gain popularity and importance. As a result, various proposals are put forward to manage videos, and one of them is data embedding. Essentially, data embedding inserts data into the video to serve a specific purpose, including proof of ownership via watermark, covert communication in steganography, and authentication via fragile watermark. However, most conventional methods embed data by using only one type of syntax element defined in the video coding standard, which may suffer from large bit rate overhead, quality degradation, or low payload. Therefore, this work aims to explore the combined use of multiple prediction syntax elements in SHVC for the purpose of data embedding. Specifically, the intra prediction mode, motion vector predictor, motion vector difference, merge mode and coding block structure are collectively manipulated to embed data. The experimental results demonstrate that, in comparison to the conventional single-venue data embedding methods, the combined use of prediction syntax elements can achieve higher payload while preserving the perceptual quality with minimal bit rate variation. In the best case scenario, a total of 556.1 kbps is embedded into the video sequence PartyScene with a drop of 0.15 dB in PSNR while experiencing a bit rate overhead of 7.4% when all prediction syntax elements are utilized altogether. A recommendation is then put forward to choose specific types of syntax element for data embedding based on the characteristics of the video.

Selected neuropeptide genes show genetic differentiation between Africans and non-Africans.

BACKGROUND: Publicly available genome data provides valuable information on the genetic variation patterns across different modern human populations. Neuropeptide genes are crucial to the nervous, immune, endocrine system, and physiological homeostasis as they play an essential role in communicating information in neuronal functions. It remains unclear how evolutionary forces, such as natural selection and random genetic drift, have affected neuropeptide genes among human populations. To date, there are over 100 known human neuropeptides from the over 1000 predicted peptides encoded in the genome. The purpose of this study is to analyze and explore the genetic variation in continental human populations across all known neuropeptide genes by examining highly differentiated SNPs between African and non-African populations. RESULTS: We identified a total of 644,225 SNPs in 131 neuropeptide genes in 6 worldwide population groups from a public database. Of these, 5163 SNPs that had ΔDAF |(African - non-African)| ≥ 0.20 were identified and fully annotated. A total of 20 outlier SNPs that included 19 missense SNPs with a moderate impact and one stop lost SNP with high impact, were identified in 16 neuropeptide genes. Our results indicate that an overall strong population differentiation was observed in the non-African populations that had a higher derived allele frequency for 15/20 of those SNPs. Highly differentiated SNPs in four genes were particularly striking: NPPA (rs5065) with high impact stop lost variant; CHGB (rs6085324, rs236150, rs236152, rs742710 and rs742711) with multiple moderate impact missense variants; IGF2 (rs10770125) and INS (rs3842753) with moderate impact missense variants that are in linkage disequilibrium. Phenotype and disease associations of these differentiated SNPs indicated their association with hypertension and diabetes and highlighted the pleiotropic effects of these neuropeptides and their role in maintaining physiological homeostasis in humans. CONCLUSIONS: We compiled a list of 131 human neuropeptide genes from multiple databases and literature survey. We detect significant population differentiation in the derived allele frequencies of variants in several neuropeptide genes in African and non-African populations. The results highlights SNPs in these genes that may also contribute to population disparities in prevalence of diseases such as hypertension and diabetes.

FACS-Based Graph Features for Real-Time Micro-Expression Recognition.

Several studies on micro-expression recognition have contributed mainly to accuracy improvement. However, the computational complexity receives lesser attention comparatively and therefore increases the cost of micro-expression recognition for real-time application. In addition, majority of the existing approaches required at least two frames (i.e., onset and apex frames) to compute features of every sample. This paper puts forward new facial graph features based on 68-point landmarks using Facial Action Coding System (FACS). The proposed feature extraction technique (FACS-based graph features) utilizes facial landmark points to compute graph for different Action Units (AUs), where the measured distance and gradient of every segment within an AU graph is presented as feature. Moreover, the proposed technique processes ME recognition based on single input frame sample. Results indicate that the proposed FACS-baed graph features achieve up to 87.33% of recognition accuracy with F1-score of 0.87 using leave one subject out cross-validation on SAMM datasets. Besides, the proposed technique computes features at the speed of 2 ms per sample on Xeon Processor E5-2650 machine.

Progressive halftone watermarking using multilayer table lookup strategy.

In this paper, a halftoning-based multilayer watermarking of low computational complexity is proposed. An additional data-hiding technique is also employed to embed multiple watermarks into the watermark to be embedded to improve the security and embedding capacity. At the encoder, the efficient direct binary search method is employed to generate 256 reference tables to ensure the output is in halftone format. Subsequently, watermarks are embedded by a set of optimized compressed tables with various textural angles for table lookup. At the decoder, the least mean square metric is considered to increase the differences among those generated phenotypes of the embedding angles and reduce the required number of dimensions for each angle. Finally, the naïve Bayes classifier is employed to collect the possibilities of multilayer information for classifying the associated angles to extract the embedded watermarks. These decoded watermarks can be further overlapped for retrieving the additional hidden-layer watermarks. Experimental results show that the proposed method requires only 8.4 ms for embedding a watermark into an image of size 512×512 , under the 32-bit Windows 7 platform running on 4GB RAM, Intel core i7 Sandy Bridge with 4GB RAM and IDE Visual Studio 2010. Finally, only 2 MB is required to store the proposed compressed reference table.

A unified data embedding and scrambling method.

Conventionally, data embedding techniques aim at maintaining high-output image quality so that the difference between the original and the embedded images is imperceptible to the naked eye. Recently, as a new trend, some researchers exploited reversible data embedding techniques to deliberately degrade image quality to a desirable level of distortion. In this paper, a unified data embedding-scrambling technique called UES is proposed to achieve two objectives simultaneously, namely, high payload and adaptive scalable quality degradation. First, a pixel intensity value prediction method called checkerboard-based prediction is proposed to accurately predict 75% of the pixels in the image based on the information obtained from 25% of the image. Then, the locations of the predicted pixels are vacated to embed information while degrading the image quality. Given a desirable quality (quantified in SSIM) for the output image, UES guides the embedding-scrambling algorithm to handle the exact number of pixels, i.e., the perceptual quality of the embedded-scrambled image can be controlled. In addition, the prediction errors are stored at a predetermined precision using the structure side information to perfectly reconstruct or approximate the original image. In particular, given a desirable SSIM value, the precision of the stored prediction errors can be adjusted to control the perceptual quality of the reconstructed image. Experimental results confirmed that UES is able to perfectly reconstruct or approximate the original image with SSIM value > 0.99 after completely degrading its perceptual quality while embedding at 7.001 bpp on average.