Maciej Zięba

Wrocław University of Science and Technology | WUT · Institute of Informatics

In computer graphics, there is a need to recover easily modifiable representations of 3D geometry and appearance from image data. We introduce a novel method for this task using 3D Gaussian Splatting, which enables intuitive scene editing through mesh adjustments. Starting with input images and camera poses, we reconstruct the underlying geometry u...

We present PPCEF, a novel method for generating probabilistically plausible counterfactual explanations (CFs). PPCEF advances beyond existing methods by combining a probabilistic formulation that leverages the data distribution with the optimization of plausibility within a unified framework. Compared to reference approaches, our method enforces pl...

Low-rank adaptation (LoRA) is a fine-tuning technique that can be applied to conditional generative diffusion models. LoRA utilizes a small number of context examples to adapt the model to a specific domain, character, style, or concept. However, due to the limited data utilized during training, the fine-tuned model performance is often characteriz...

Enhancing low-light images with natural colors poses a challenge due to camera processing variations and limited access to ground-truth lighting conditions. To address this, we propose Dimma, a semi-supervised approach that aligns with any camera using a small set of image pairs captured under extreme lighting conditions. Our method employs a convo...

We introduce NodeFlow, a flexible framework for probabilistic regression on tabular data that combines Neural Oblivious Decision Ensembles (NODEs) and Conditional Continuous Normalizing Flows (CNFs). It offers improved modeling capabilities for arbitrary probabilistic distributions, addressing the limitations of traditional parametric approaches. I...

We introduce NodeFlow, a flexible framework for probabilistic regression on tabular data that combines Neural Oblivious Decision Ensemble (NODE) and Conditional Continuous Normalizing Flows (CNF). It offers improved modeling capabilities for arbitrary probabilistic distributions, addressing the limitations of traditional parametric approaches. In N...

We present PPCEF, a novel method for generating probabilistically plausible counterfactual explanations (CFs). PPCEF advances beyond existing methods by combining a probabilistic formulation that leverages the data distribution with the optimization of plausibility within a unified framework. Compared to reference approaches, our method enforces pl...

Growing regulatory and societal pressures demand increased transparency in AI, particularly in understanding the decisions made by complex machine learning models. Counterfactual Explanations (CFs) have emerged as a promising technique within Explainable AI (xAI), offering insights into individual model predictions. However, to understand the syste...

Although modern generative models achieve excellent quality in a variety of tasks, they often lack the essential ability to generate examples with requested properties, such as the age of the person in the photo or the weight of the generated molecule. To overcome these limitations we propose PluGeN (Plugin Generative Network), a simple yet effecti...

Designing predictive models for subjective problems in natural language processing (NLP) remains challenging. This is mainly due to its non-deterministic nature and different perceptions of the content by different humans. It may be solved by Personalized Natural Language Processing (PNLP), where the model exploits additional information about the...

Estimation of patient-specific hemodynamic features, and in particular fractional flow reserve (FFR) in coronary arteries is an essential step in providing personalized and accurate diagnosis of coronary artery disease (CAD). In recent years, in the domain of computed tomography angiography (CTA), a virtual FFR (vFFR) derived from coronary CTA usin...

The tree-based ensembles are known for their outstanding performance in classification and regression problems characterized by feature vectors represented by mixed-type variables from various ranges and domains. However, considering regression problems, they are primarily designed to provide deterministic responses or model the uncertainty of the...

Self-supervised methods have been proven effective for learning deep representations of 3D point cloud data. Although recent methods in this domain often rely on random masking of inputs, the results of this approach can be improved. We introduce PointCAM, a novel adversarial method for learning a masking function for point clouds. Our model utiliz...

Few-shot models aim at making predictions using a minimal number of labeled examples from a given task. The main challenge in this area is the one-shot setting, where only one element represents each class. We propose the general framework for few-shot learning via kernel HyperNetworks—the fusion of kernels and hypernetwork paradigm. Firstly, we in...

NeRF is a popular model that efficiently represents 3D objects from 2D images. However, vanilla NeRF has a few important limitations. NeRF must be trained on each object separately. The training time is long since we encode the object's shape and color in neural network weights. Moreover, NeRF does not generalize well to unseen data. In this paper,...

Traditional 3D face models are based on mesh representations with texture. One of the most important models is FLAME (Faces Learned with an Articulated Model and Expressions), which produces meshes of human faces that are fully controllable. Unfortunately, such models have problems with capturing geometric and appearance details. In contrast to mes...

Recently, generative models for 3D objects are gaining much popularity in VR and augmented reality applications. Training such models using standard 3D representations, like voxels or point clouds, is challenging and requires complex tools for proper color rendering. In order to overcome this limitation, Neural Radiance Fields (NeRFs) offer a state...

Talking face generation has historically struggled to produce head movements and natural facial expressions without guidance from additional reference videos. Recent developments in diffusion-based generative models allow for more realistic and stable data synthesis and their performance on image and video generation has surpassed that of other gen...

Iris-based identification systems are among the most popular approaches for person identification. Such systems require good-quality segmentation modules that ideally identify the regions for different eye components. This paper introduces the new two-headed architecture, where the eye components and eyelashes are segmented using two separate decod...

Modern generative models achieve excellent quality in a variety of tasks including image or text generation and chemical molecule modeling. However, existing methods often lack the essential ability to generate examples with requested properties, such as the age of the person in the photo or the weight of the generated molecule. Incorporating such...

The tree-based ensembles are known for their outstanding performance for classification and regression problems characterized by feature vectors represented by mixed-type variables from various ranges and domains. However, considering regression problems, they are primarily designed to provide deterministic responses or model the uncertainty of the...

The aim of Few-Shot learning methods is to train models which can easily adapt to previously unseen tasks, based on small amounts of data. One of the most popular and elegant Few-Shot learning approaches is Model-Agnostic Meta-Learning (MAML). The main idea behind this method is to learn the general weights of the meta-model, which are further adap...

Contemporary deep neural networks offer state-of-the-art results when applied to visual reasoning, e.g., in the context of 3D point cloud data. Point clouds are important datatype for precise modeling of three-dimensional environments, but effective processing of this type of data proves to be challenging. In the world of large, heavily-parameteriz...

Few-shot models aim at making predictions using a minimal number of labeled examples from a given task. The main challenge in this area is the one-shot setting where only one element represents each class. We propose HyperShot - the fusion of kernels and hypernetwork paradigm. Compared to reference approaches that apply a gradient-based adjustment...

In this work, we propose a novel method for generating 3D point clouds that leverages properties of hypernetworks. Contrary to the existing methods that learn only the representation of a 3D object, our approach simultaneously finds a representation of the object and its 3D surface. The main idea of our HyperCloud method is to build a hypernetwork...

Gaussian Processes (GPs) have been widely used in machine learning to model distributions over functions, with applications including multi-modal regression, time-series prediction, and few-shot learning. GPs are particularly useful in the last application since they rely on Normal distributions and enable closed-form computation of the posterior p...

Generative models have gained many researchers' attention in the last years resulting in models such as StyleGAN for human face generation or PointFlow for the 3D point cloud generation. However, by default, we cannot control its sampling process, i.e., we cannot generate a sample with a specific set of attributes. The current approach is model ret...

In this paper, we propose a simple yet effective method to represent point clouds as sets of samples drawn from a cloud-specific probability distribution. This interpretation matches intrinsic characteristics of point clouds: the number of points and their ordering within a cloud is not important as all points are drawn from the proximity of the ob...

Modern generative models achieve excellent quality in a variety of tasks including image or text generation and chemical molecule modeling. However, existing methods often lack the essential ability to generate examples with requested properties, such as the age of the person in the photo or the weight of the generated molecule. Incorporating such...

Recently proposed 3D object reconstruction methods represent a mesh with an atlas - a set of planar patches approximating the surface. However, their application in a real-world scenario is limited since the surfaces of reconstructed objects contain discontinuities, which degrades the quality of the final mesh. This is mainly caused by independent...

Predicting future states or actions of a given system remains a fundamental, yet unsolved challenge of intelligence, especially in the scope of complex and non-deterministic scenarios, such as modeling behavior of humans. Existing approaches provide results under strong assumptions concerning unimodality of future states, or, at best, assuming spec...

In this paper, we propose a simple yet effective method to represent point clouds as sets of samples drawn from a cloud-specific probability distribution. This interpretation matches intrinsic characteristics of point clouds: the number of points and their ordering within a cloud is not important as all points are drawn from the proximity of the ob...

The three-dimensional data representations have found numerous applications, most notably in SLAM and autonomous driving, where the most widely used type of data is a point cloud. In contrast to the image data, point clouds are unstructured objects, represented as sets of points in three- or six-dimensional (if the colors of surroundings are captur...

Signed distance field (SDF) is a prominent implicit representation of 3D meshes. Methods that are based on such representation achieved state-of-the-art 3D shape reconstruction quality. However, these methods struggle to reconstruct non-convex shapes. One remedy is to incorporate a constructive solid geometry framework (CSG) that represents a shape...

In this work, we present HyperFlow - a novel generative model that leverages hypernetworks to create continuous 3D object representations in a form of lightweight surfaces (meshes), directly out of point clouds. Efficient object representations are essential for many computer vision applications, including robotic manipulation and autonomous drivin...

The three-dimensional data is the core tool behind environment aware algorithms used in e.g. SLAM or autonomous driving. As a data format, point clouds are becoming increasingly popular, due to their high-resolution and mapping fidelity. However, representing data as points, rather than voxels, comes with very high processing complexity, as machine...

In this work, we perform a comprehensive study on post-training quantization methods for convolutional neural networks in two challenging tasks: classification and object detection. Furthermore, we introduce a novel method that quantizes every single layer to the smallest bit width, which does not introduce accuracy degradation. As a result, the mo...

The recent development in the fields of autonomous vehicles, robot vision and virtual reality caused a shift in the research focus - more attention is paid to 3D data representation. In this work, we introduce a novel approach for learning representations for 3D point clouds in semi-supervised mode. The main idea of the approach is to combine the b...

In this work, we propose a novel method for generating 3D point clouds that leverage properties of hyper networks. Contrary to the existing methods that learn only the representation of a 3D object, our approach simultaneously finds a representation of the object and its 3D surface. The main idea of our HyperCloud method is to build a hyper network...

Deep generative architectures provide a way to model not only images but also complex, 3-dimensional objects, such as point clouds. In this work, we present a novel method to obtain meaningful representations of 3D shapes that can be used for challenging tasks, including 3D points generation, reconstruction, compression, and clustering. Contrary to...

This paper focuses on a novel generative approach for 3D point clouds that makes use of invertible flow-based models. The main idea of the method is to treat a point cloud as a probability density in 3D space that is modeled using a cloud-specific neural network. To capture the similarity between point clouds we rely on parameter sharing among netw...

In traditional generative modeling, good data representation is very often a base for a good machine learning model. It can be linked to good representations encoding more explanatory factors that are hidden in the original data. With the invention of Generative Adversarial Networks (GANs), a subclass of generative models that are able to learn rep...

In traditional generative modeling, good data representation is very often a base for a good machine learning model. It can be linked to good representations encoding more explanatory factors that are hidden in the original data. With the invention of Generative Adversarial Networks (GANs), a subclass of generative models that are able to learn rep...

In this work we introduce a novel approach to train Bidirectional Generative Adversarial Model (BiGAN) in a semi-supervised manner. The presented method utilizes triplet loss function as an additional component of the objective function used to train discriminative data representation in the latent space of the BiGAN model. This representation can...

In this work we introduce a novel approach to train Bidirectional Generative Adversarial Model (BiGAN) in a semi-supervised manner. The presented method utilizes triplet loss function as an additional component of the objective function used to train discriminative data representation in the latent space of the BiGAN model. This representation can...

Deep generative architectures provide a way to model not only images but also complex, 3-dimensional objects, such as point clouds. In this work, we present a novel method to obtain meaningful representations of 3D shapes that can be used for challenging tasks including 3D points generation, reconstruction, compression, and clustering. Contrary to...

In this paper, we propose a novel regularization method for Generative Adversarial Networks, which allows the model to learn discriminative yet compact binary representations of image patches (image descriptors). We employ the dimensionality reduction that takes place in the intermediate layers of the discriminator network and train binarized low-d...

Tracking of signals in Nuclear Magnetic Resonance (NMR) spectra is a basic technique used in drug discovery, systems and structural biology. Current experimental setups allow to measure hundreds of spectra, which require analysis, ideally in an automated and reproducible manner. In this study, we present a novel approach to the automate tracking of...

Motivation: Automated selection of signals in protein NMR spectra, known as peak picking, has been studied for over 20 years, nevertheless existing peak picking methods are still largely deficient. Accurate and precise automated peak picking would accelerate the structure calculation, and analysis of dynamics and interactions of macromolecules. Re...

Triplet networks are widely used models that are characterized by good performance in classification and retrieval tasks. In this work we propose to train a triplet network by putting it as the discriminator in Generative Adversarial Nets (GANs). We make use of the good capability of representation learning of the discriminator to increase the pred...

In this paper, we propose a novel training paradigm that combines two learning strategies: cost-sensitive and self-paced learning. This learning approach can be applied to the decision problems where highly imbalanced data is used during training process. The main idea behind the proposed method is to start the learning process by taking large numb...

In this work we present a novel ensemble model for a credit scoring problem. The main idea of the approach is to incorporate separate beta binomial distributions for each of the classes to generate balanced datasets that are further used to construct base learners that constitute the final ensemble model. The sampling procedure is performed on two...

In this work we present a novel ensemble model for a credit scoring problem. The main idea of the approach is to incorporate separate beta binomial distributions for each of the classes to generate balanced datasets that are further used to construct base learners that constitute the final ensemble model. The sampling procedure is performed on two...

Application of machine learning to medical diagnosis entails facing two major issues, namely, a necessity of learning comprehensible models and a need of coping with imbalanced data phenomenon. The first one corresponds to a problem of implementing interpretable models, e.g., classification rules or decision trees. The second issue represents a sit...

The problem of imbalanced data, i.e., when the class labels are unequally distributed, is encountered in many real-life application, e.g., credit scoring, medical diagnostics. Various approaches aimed at dealing with the imbalanced data have been proposed. One of the most well known data pre-processing method is the Synthetic Minority Oversampling...

The problem of selecting the proper set of questions plays very important role in the domain of information retrieval, e.g., on the Internet, or about requirements during the business interview. In this work we propose a novel approach for selecting the sequence of binary questions to be asked to identify an unknown concept. This solution makes use...

Credit scoring is the assessment of the risk associated with a consumer (an organization or an individual) that apply for the credit. Therefore, the problem of credit scoring can be stated as a discrimination between those applicants whom the lender is confident will repay credit and those applicants who are considered by the lender as insufficient...

In this work, we introduce a novel training method for constructing boosted Support Vector Machines (SVMs) directly from imbalanced data. The proposed solution incorporates the mechanisms of active learning strategy to eliminate redundant instances and more properly estimate misclassification costs for each of the base SVMs in the committee. To eva...

In this work we propose a service-oriented support decision system (SOSDS) for diagnostic problems that is insensitive to the problems of the imbalanced data and missing values of the attributes, which are widely observed in medical domain. The system is composed of distributed Web services, which implement machine learning solutions dedicated to c...

In this paper we propose a novel combined approach to solve the imbalanced data issue in the application to the problem of the post-operative life expectancy prediction for the lung cancer patients. This solution makes use of undersampling techniques together with cost-sensitive SVM (Support Vector Machines). First, we eliminate non-informative exa...

In this work, we propose the ensemble SVM that solves the problem of missing values of attributes and the imbalanced data phenomenon in the domain of postoperative risk management. Contrary to the other approaches the our solution effectively deals with the problems of high percentage of unknown values of the features. The problem of imbalanced dat...

In real-life situations characteristics of Web service systems evolve in time. Therefore, change detection techniques become substantial elements of adaptive procedures for Web service systems management, such as resource allocation and anomaly detection methods. In this paper, we propose an on-line change detector which uses the Bayesian inference...

The goal of this paper is to propose an ensemble classification method for the credit assignment problem. The idea of the proposed method is based on switching class labels techniques. An application of such techniques allows solving two typical data mining problems: a predicament of imbalanced dataset, and an issue of asymmetric cost matrix. The p...

The aim of the paper is to discuss some selected issues related to services merging, partitioning and execution in systems based on service oriented paradigm. The main feature of such systems is that the required services may be efficiently and flexibly composed of available atomic (elementary) services providing certain and well-defined functional...

This paper presents preliminary studies on the problem of classification of different kinds of human arm motions based on EMG signals. Methods of change detection, classification, features extraction and selection are considered as an important elements of recognition process. Presented algorithms are part of module to visualise human arm movements...

There are various approaches to solve the problem of services selection according to user request. Usually, ontology-based filtering is used. The problem occurs if the service can be run a limited number of times and the number of users interested in service execution is higher than the number of possible service executions. In this work we propose...

In this work we propose an innovative approach to data mining problem. We propose very flexible data mining system based on service-oriented architecture. Developing applications according to SOA paradigm emerges from the rapid development of the new technology direct known as sustainability science. Each of data mining functionalities is delivered...

In this work the proposal for services recommendation in online educational systems based on service oriented architecture is introduced. The problem of recommending services responsible for creating student groups are taken into account and as the criterion of the grouping the student learning potential is considered. As a method of grouping modif...