Rectified Flow (RF) has been widely used as an effective generative model. Although RF is primarily based on probability flow Ordinary Differential Equations (ODE), recent studies have shown that injecting noise through reverse-time Stochastic Differential Equations (SDE) for sampling can achieve superior generative performance. Inspired by Positive-incentive Noise (Pi-noise), we propose an innovative generative algorithm to train Pi-noise generators, namely Rectified Noise (RN), which improves the generative performance by injecting Pi-noise into the velocity field of pre-trained RF models. After introducing the Rectified Noise pipeline, pre-trained RF models can be efficiently transformed into Pi-noise generators. We validate Rectified Noise by conducting extensive experiments across various model architectures on different datasets. Notably, we find that: (1) RF models using Rectified Noise reduce FID from10.16 to 9.05 on ImageNet-1k. (2) The models of Pi-noise generators achieve improved performance with only 0.39% additional training parameters.

A large number of works aim to alleviate the impact of noise due to an underlying conventional assumption of the negative role of noise. However, some existing works show that the assumption does not always hold. In this paper, we investigate how to benefit the classical models by random noise under the framework of Positive-incentive Noise (Pi-Noise) (Li, 2024). Since the ideal objective of Pi-Noise is intractable, we propose to optimize its variational bound instead, namely variational Pi-Noise (VPN). With the variational inference, a VPN generator implemented by neural networks is designed for enhancing base models and simplifying the inference of base models, without changing the architecture of base models. Benefiting from the independent design of base models and VPN generators, the VPN generator can work with most existing models. From the extensive experiments on different base models (including linear models, ResNet, ViT, etc.) it is shown that the proposed VPN generator can improve the base models. It is appealing that the trained VPN generator prefers to blur the irrelevant ingredients in complicated images, which meets our expectations.

Cross-spectrum depth estimation aims to provide a reliable depth map under variant-illumination conditions with a pair of dual-spectrum images. It is valuable for autonomous driving applications when vehicles are equipped with two cameras of different modalities. However, images captured by different-modality cameras can be photometrically quite different, which makes cross-spectrum depth estimation a very challenging problem. Moreover, the shortage of large-scale open-source datasets also retards further research in this field. In this paper, we propose an unsupervised visible light(VIS)-image-guided cross-spectrum (i.e., thermal and visible-light, TIR-VIS in short) depth-estimation framework. The input of the framework consists of a cross-spectrum stereo pair (one VIS image and one thermal image). First, we train a depth-estimation base network using VIS-image stereo pairs. To adapt the trained depth-estimation network to the cross-spectrum images, we propose a multi-scale feature-transfer network to transfer features from the TIR domain to the VIS domain at the feature level. Furthermore, we introduce a mechanism of cross-spectrum depth cycle-consistency to improve the depth estimation result of dual-spectrum image pairs. Meanwhile, we release to society a large cross-spectrum dataset with visible-light and thermal stereo images captured in different scenes. The experiment result shows that our method achieves better depth-estimation results than the compared existing methods. Our code and dataset are available onhttps://github.com/whitecrow1027/CrossSP_Depth.

In this work, we propose a novel paradigm to design a hybrid thermal/EO (Electro-Optical or visible-light) camera, whose thermal and RGB frames are pixel-wisely aligned and temporally synchronized. Compared with the existing schemes, we innovate in three ways in order to make it more compact in dimension, and thus more practical and extendable for real-world applications. The first is a redesign of the structure layout of the thermal and EO cameras. The second is on obtaining a pixel-wise spatial registration of the thermal and RGB frames by a coarse mechanical adjustment and a fine alignment through a constant homography warping. The third innovation is on extending one single hybrid camera to a hybrid camera array, through which we can obtain wide-view spatially aligned thermal, RGB and disparity images simultaneously. The experimental results show that the average error of spatial-alignment of two image modalities can be less than one pixel.

MOTIVATION: Protein post-translational modifications (PTMs) regulate a wide range of cellular protein functions. Many PTM sites from the same (intra) or different (inter) proteins often cooperate with each other to perform a function, which is defined as PTM cross-talk. PTM cross-talk within proteins attracted great attentions in the past a few years. However, the inter-protein PTM cross-talk is largely under studied due to its large protein pair space and lack of a gold standard dataset, even though the PTM interplay between proteins is a key element in cell signaling and regulatory networks. RESULTS: In this study, 199 inter-protein PTM cross-talk pairs in 82 pairs of human proteins were collected from literature, which to our knowledge is the first effort in compiling such dataset. By comparing with background PTM pairs from the same protein pairs, we found that inter-protein cross-talk PTM pairs have higher sequence co-evolution at both PTM residue and motif levels. Also, we found that cross-talk PTMs have higher co-modification across multiple species and 88 human tissues or conditions. Furthermore, we showed that these features are predictive for PTM cross-talk between proteins, and applied a random forest model to integrate these features with achieving an area under the receiver operating characteristic curve of 0.81 in 10-fold cross-validation, prevailing over using any single feature alone. Therefore, this method would be a valuable tool to identify inter-protein PTM cross-talk at proteome-wide scale. AVAILABILITY AND IMPLEMENTATION: A web server for prioritization of both intra- and inter-protein PTM cross-talk candidates is at http://bioinfo.bjmu.edu.cn/ptm-x/. Python code for local computer is also freely available at https://github.com/huangyh09/PTM-X. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Analyzing the popularity of microblogging services is of great significance in various applications. The number of posts provides novel insights into the popularity of microblogging services, and is critical to learn about the frequency of use. Existing approaches analyze this parameter by observing posts published by a large number of users, which may lead to underestimate the value since the sampled user may stop to use the service during the observing process. In this paper, we propose a novel method to estimate the number of posts in microblogging services. The basic idea behind this method is to make use of a common API provided by microblogging services, i.e., the public_timeline API. Posts sampled by this API may duplicate among multiple invocations, so the capture-recapture model can be used to estimate the total number of posts. Based on the traditional capture-recapture model, we propose an improved model to address challenges on low sampling probability and unequal sampling probability. We validate the proposed method using a real life Sina Weibo dataset, and the experimental results demonstrate the effectiveness and accuracy of our proposed method.

In this paper the concept of fuzzy IBE schemes with some fixed attributes (SAA-FIBE) is proposed and one construction of it is presented. SAA-FIBE scheme can be viewed as a variant of SW scheme described in which demanding no fixed positive or negative attributes. In our scheme, a user with identity omega can decrypt the message that is encrypted with a set of attributes, omega', if and only if |omega' capomega| ges d and omega must have or must have not some attributes described in encryption policy. The scheme are both error-tolerant and secure against collusion attacks in the SPID-FIBE attack model.