Compositional Text-to-Image Synthesis with Attention Map Control of Diffusion Models

Recent text-to-image (T2I) diffusion models show outstanding performance in generating high-quality images conditioned on textual prompts. However, these models fail to semantically align the generated images with the text descriptions due to their limited compositional capabilities, leading to attribute leakage, entity leakage, and missing entities. In this paper, we propose a novel attention mask control strategy based on predicted object boxes to address these three issues. In particular, we first train a BoxNet to predict a box for each entity that possesses the attribute specified in the prompt. Then, depending on the predicted boxes, unique mask control is applied to the cross- and self-attention maps. Our approach produces a more semantically accurate synthesis by constraining the attention regions of each token in the prompt to the image. In addition, the proposed method is straightforward and effective, and can be readily integrated into existing cross-attention-diffusion-based T2I generators. We compare our approach to competing methods and demonstrate that it not only faithfully conveys the semantics of the original text to the generated content, but also achieves high availability as a ready-to-use plugin.

Edit Everything: A Text-Guided Generative System for Images Editing

We introduce a new generative system called Edit Everything, which can take image and text inputs and produce image outputs. Edit Everything allows users to edit images using simple text instructions. Our system designs prompts to guide the visual module in generating requested images. Experiments demonstrate that Edit Everything facilitates the implementation of the visual aspects of Stable Diffusion with the use of Segment Anything model and CLIP. Our system is publicly available at https://github.com/DefengXie/Edit_Everything.

UNADON: Transformer-based model to predict genome-wide chromosome spatial position

The spatial positioning of chromosomes relative to functional nuclear bodies is intertwined with genome functions such as transcription. However, the sequence patterns and epigenomic features that collectively influence chromatin spatial positioning in a genome-wide manner are not well understood. Here, we develop a new transformer-based deep learning model called UNADON, which predicts the genome-wide cytological distance to a specific type of nuclear body, as measured by TSA-seq, using both sequence features and epigenomic signals. Evaluations of UNADON in four cell lines (K562, H1, HFFc6, HCT116) show high accuracy in predicting chromatin spatial positioning to nuclear bodies when trained on a single cell line. UNADON also performed well in an unseen cell type. Importantly, we reveal potential sequence and epigenomic factors that affect large-scale chromatin compartmentalization to nuclear bodies. Together, UNADON provides new insights into the principles between sequence features and large-scale chromatin spatial localization, which has important implications for understanding nuclear structure and function.

System Identification with Copula Entropy

Identifying differential equation governing dynamical system is an important problem with wide applications. Copula Entropy (CE) is a mathematical concept for measuring statistical independence in information theory. In this paper we propose a method for identifying differential equation of dynamical systems with CE. The problem is considered as a variable selection problem and solved with the previously proposed CE-based method for variable selection. The proposed method composed of two components: the difference operator and the CE estimator. Since both components can be done non-parametrically, the proposed method is therefore model-free and hyperparameter-free. The simulation experiment with the 3D Lorenz system verified the effectiveness of the proposed method.

GlyphDraw: Learning to Draw Chinese Characters in Image Synthesis Models Coherently

Recent breakthroughs in the field of language-guided image generation have yielded impressive achievements, enabling the creation of high-quality and diverse images based on user instructions. Although the synthesis performance is fascinating, one significant limitation of current image generation models is their insufficient ability to generate coherent text within images, particularly for complex glyph structures like Chinese characters. To address this problem, we introduce GlyphDraw, a general learning framework aiming at endowing image generation models with the capacity to generate images embedded with coherent text. To the best of our knowledge, this is the first work in the field of image synthesis to address the generation of Chinese characters. % we first adopt the OCR technique to collect images with Chinese characters as training samples, and extract the text and locations as auxiliary information. We first sophisticatedly design the image-text dataset's construction strategy, then build our model specifically on a diffusion-based image generator and carefully modify the network structure to allow the model to learn drawing Chinese characters with the help of glyph and position information. Furthermore, we maintain the model's open-domain image synthesis capability by preventing catastrophic forgetting by using a variety of training techniques. Extensive qualitative and quantitative experiments demonstrate that our method not only produces accurate Chinese characters as in prompts, but also naturally blends the generated text into the background. Please refer to https://1073521013.github.io/glyph-draw.github.io

Identifying Time Lag in Dynamical Systems with Copula Entropy based Transfer Entropy

Time lag between variables is a key characteristics of dynamical systems in different fields and identifying such time lag is a central problem in complex systems with many applications. Transfer Entropy (TE) was proposed as a tool for time lag identification recently. Unfortunately, estimating TE has been a notoriously difficult problem. Copula Entropy (CE) is a measure of statistical independence and it was proved that TE can be represented with only CE. Therefore, a non-parametric estimator of TE based on CE was proposed according to such representation recently. In this paper we propose to use the CE-based estimator of TE to identify time lag in dynamical systems. Both simulated and real data are used to verify the effectiveness of the proposed method in the experiments. Experimental results show that the proposed method can identify the time lags in the three simulated systems. The real data experiment with the data on power consumption of the Tetouan city also demonstrates that our method can identify the pattern of time lags through the estimated TE from the weather factors to the power consumption of the city.

EPIC-KITCHENS VISOR Benchmark: VIdeo Segmentations and Object Relations

We introduce VISOR, a new dataset of pixel annotations and a benchmark suite for segmenting hands and active objects in egocentric video. VISOR annotates videos from EPIC-KITCHENS, which comes with a new set of challenges not encountered in current video segmentation datasets. Specifically, we need to ensure both short- and long-term consistency of pixel-level annotations as objects undergo transformative interactions, e.g. an onion is peeled, diced and cooked - where we aim to obtain accurate pixel-level annotations of the peel, onion pieces, chopping board, knife, pan, as well as the acting hands. VISOR introduces an annotation pipeline, AI-powered in parts, for scalability and quality. In total, we publicly release 272K manual semantic masks of 257 object classes, 9.9M interpolated dense masks, 67K hand-object relations, covering 36 hours of 179 untrimmed videos. Along with the annotations, we introduce three challenges in video object segmentation, interaction understanding and long-term reasoning. For data, code and leaderboards: http://epic-kitchens.github.io/VISOR

Copula Entropy based Variable Selection for Survival Analysis

Variable selection is an important problem in statistics and machine learning. Copula Entropy (CE) is a mathematical concept for measuring statistical independence and has been applied to variable selection recently. In this paper we propose to apply the CE-based method for variable selection to survival analysis. The idea is to measure the correlation between variables and time-to-event with CE and then select variables according to their CE value. Experiments on simulated data and two real cancer data were conducted to compare the proposed method with two related methods: random survival forest and Lasso-Cox. Experimental results showed that the proposed method can select the 'right' variables out that are more interpretable and lead to better prediction performance.

Subband-based Generative Adversarial Network for Non-parallel Many-to-many Voice Conversion

Voice conversion is to generate a new speech with the source content and a target voice style. In this paper, we focus on one general setting, i.e., non-parallel many-to-many voice conversion, which is close to the real-world scenario. As the name implies, non-parallel many-to-many voice conversion does not require the paired source and reference speeches and can be applied to arbitrary voice transfer. In recent years, Generative Adversarial Networks (GANs) and other techniques such as Conditional Variational Autoencoders (CVAEs) have made considerable progress in this field. However, due to the sophistication of voice conversion, the style similarity of the converted speech is still unsatisfactory. Inspired by the inherent structure of mel-spectrogram, we propose a new voice conversion framework, i.e., Subband-based Generative Adversarial Network for Voice Conversion (SGAN-VC). SGAN-VC converts each subband content of the source speech separately by explicitly utilizing the spatial characteristics between different subbands. SGAN-VC contains one style encoder, one content encoder, and one decoder. In particular, the style encoder network is designed to learn style codes for different subbands of the target speaker. The content encoder network can capture the content information on the source speech. Finally, the decoder generates particular subband content. In addition, we propose a pitch-shift module to fine-tune the pitch of the source speaker, making the converted tone more accurate and explainable. Extensive experiments demonstrate that the proposed approach achieves state-of-the-art performance on VCTK Corpus and AISHELL3 datasets both qualitatively and quantitatively, whether on seen or unseen data. Furthermore, the content intelligibility of SGAN-VC on unseen data even exceeds that of StarGANv2-VC with ASR network assistance.

Evaluating Independence and Conditional Independence Measures

Independence and Conditional Independence (CI) are two fundamental concepts in probability and statistics, which can be applied to solve many central problems of statistical inference. There are many existing independence and CI measures defined from diverse principles and concepts. In this paper, the 16 independence measures and 16 CI measures were reviewed and then evaluated with simulated and real data. For the independence measures, eight simulated data were generating from normal distribution, normal and Archimedean copula functions to compare the measures in bivariate or multivariate, linear or nonlinear settings. Two UCI dataset, including the heart disease data and the wine quality data, were used to test the power of the independence measures in real conditions. For the CI measures, two simulated data with normal distribution and Gumbel copula, and one real data (the Beijing air data) were utilized to test the CI measures in prespecified linear or nonlinear setting and real scenario. From the experimental results, we found that most of the measures work well on the simulated data by presenting the right monotonicity of the simulations. However, the independence and CI measures were differentiated on much complex real data respectively and only a few can be considered as working well with reference to domain knowledge. We also found that the measures tend to be separated into groups based on the similarity of the behaviors of them in each setting and in general. According to the experiments, we recommend CE as a good choice for both independence and CI measure. This is also due to its rigorous distribution-free definition and consistent nonparametric estimator.