arXiv subjects

Tomasz Trzcinski

Publications and source records attributed to Tomasz Trzcinski.

At least 19 recordsLinked to original sources

arXiv2025

Neural Bloom: A Deep Learning Approach to Real-Time Lighting

We propose a novel method to generate bloom lighting effect in real time using neural networks. Our solution generate brightness mask from given 3D scene view up to 30% faster than state-of-the-art methods. The existing traditional techniques rely on multiple blur appliances and texture sampling, also very often have existing conditional branching in its implementation. These operations occupy big portion of the execution time. We solve this problem by proposing two neural network-based bloom lighting methods, Neural Bloom Lighting (NBL) and Fast Neural Bloom Lighting (FastNBL), focusing on their quality and performance. Both methods were tested on a variety of 3D scenes, with evaluations conducted on brightness mask accuracy and inference speed. The main contribution of this work is that both methods produce high-quality bloom effects while outperforming the standard state-of-the-art bloom implementation, with FastNBL being faster by 28% and NBL faster by 12%. These findings highlight that we can achieve realistic bloom lighting phenomena faster, moving us towards more realism in real-time environments in the future. This improvement saves computational resources, which is a major bottleneck in real-time rendering. Furthermore, it is crucial for sustaining immersion and ensuring smooth experiences in high FPS environments, while maintaining high-quality realism.

cs.CV↗

arXiv2024

LumiGauss: Relightable Gaussian Splatting in the Wild

Decoupling lighting from geometry using unconstrained photo collections is notoriously challenging. Solving it would benefit many users as creating complex 3D assets takes days of manual labor. Many previous works have attempted to address this issue, often at the expense of output fidelity, which questions the practicality of such methods. We introduce LumiGauss - a technique that tackles 3D reconstruction of scenes and environmental lighting through 2D Gaussian Splatting. Our approach yields high-quality scene reconstructions and enables realistic lighting synthesis under novel environment maps. We also propose a method for enhancing the quality of shadows, common in outdoor scenes, by exploiting spherical harmonics properties. Our approach facilitates seamless integration with game engines and enables the use of fast precomputed radiance transfer. We validate our method on the NeRF-OSR dataset, demonstrating superior performance over baseline methods. Moreover, LumiGauss can synthesize realistic images for unseen environment maps. Our code: https://github.com/joaxkal/lumigauss.

cs.CV↗

arXiv2023

Noninvasive Estimation of Mean Pulmonary Artery Pressure Using MRI, Computer Models, and Machine Learning

Pulmonary Hypertension (PH) is a severe disease characterized by an elevated pulmonary artery pressure. The gold standard for PH diagnosis is measurement of mean Pulmonary Artery Pressure (mPAP) during an invasive Right Heart Catheterization. In this paper, we investigate noninvasive approach to PH detection utilizing Magnetic Resonance Imaging, Computer Models and Machine Learning. We show using the ablation study, that physics-informed feature engineering based on models of blood circulation increases the performance of Gradient Boosting Decision Trees-based algorithms for classification of PH and regression of values of mPAP. We compare results of regression (with thresholding of estimated mPAP) and classification and demonstrate that metrics achieved in both experiments are comparable. The predicted mPAP values are more informative to the physicians than the probability of PH returned by classification models. They provide the intuitive explanation of the outcome of the machine learning model (clinicians are accustomed to the mPAP metric, contrary to the PH probability).

eess.IV↗

arXiv2023

Learning Data Representations with Joint Diffusion Models

Joint machine learning models that allow synthesizing and classifying data often offer uneven performance between those tasks or are unstable to train. In this work, we depart from a set of empirical observations that indicate the usefulness of internal representations built by contemporary deep diffusion-based generative models not only for generating but also predicting. We then propose to extend the vanilla diffusion model with a classifier that allows for stable joint end-to-end training with shared parameterization between those objectives. The resulting joint diffusion model outperforms recent state-of-the-art hybrid methods in terms of both classification and generation quality on all evaluated benchmarks. On top of our joint training approach, we present how we can directly benefit from shared generative and discriminative representations by introducing a method for visual counterfactual explanations.

cs.LG↗

arXiv2022

A Deep Learning Approach for Automatic Detection of Qualitative Features of Lecturing

Artificial Intelligence in higher education opens new possibilities for improving the lecturing process, such as enriching didactic materials, helping in assessing students' works or even providing directions to the teachers on how to enhance the lectures. We follow this research path, and in this work, we explore how an academic lecture can be assessed automatically by quantitative features. First, we prepare a set of qualitative features based on teaching practices and then annotate the dataset of academic lecture videos collected for this purpose. We then show how these features could be detected automatically using machine learning and computer vision techniques. Our results show the potential usefulness of our work.

cs.AI↗

arXiv2021

EgoNN: Egocentric Neural Network for Point Cloud Based 6DoF Relocalization at the City Scale

The paper presents a deep neural network-based method for global and local descriptors extraction from a point cloud acquired by a rotating 3D LiDAR. The descriptors can be used for two-stage 6DoF relocalization. First, a course position is retrieved by finding candidates with the closest global descriptor in the database of geo-tagged point clouds. Then, the 6DoF pose between a query point cloud and a database point cloud is estimated by matching local descriptors and using a robust estimator such as RANSAC. Our method has a simple, fully convolutional architecture based on a sparse voxelized representation. It can efficiently extract a global descriptor and a set of keypoints with local descriptors from large point clouds with tens of thousand points. Our code and pretrained models are publicly available on the project website.

cs.CV↗

arXiv2021

Large-Scale Topological Radar Localization Using Learned Descriptors

In this work, we propose a method for large-scale topological localization based on radar scan images using learned descriptors. We present a simple yet efficient deep network architecture to compute a rotationally invariant discriminative global descriptor from a radar scan image. The performance and generalization ability of the proposed method is experimentally evaluated on two large scale driving datasets: MulRan and Oxford Radar RobotCar. Additionally, we present a comparative evaluation of radar-based and LiDAR-based localization using learned global descriptors. Our code and trained models are publicly available on the project website.

cs.CV↗

arXiv2021

I Want This Product but Different : Multimodal Retrieval with Synthetic Query Expansion

This paper addresses the problem of media retrieval using a multimodal query (a query which combines visual input with additional semantic information in natural language feedback). We propose a SynthTriplet GAN framework which resolves this task by expanding the multimodal query with a synthetically generated image that captures semantic information from both image and text input. We introduce a novel triplet mining method that uses a synthetic image as an anchor to directly optimize for embedding distances of generated and target images. We demonstrate that apart from the added value of retrieval illustration with synthetic image with the focus on customization and user feedback, the proposed method greatly surpasses other multimodal generation methods and achieves state of the art results in the multimodal retrieval task. We also show that in contrast to other retrieval methods, our method provides explainable embeddings.

cs.CV↗

arXiv2021

MinkLoc++: Lidar and Monocular Image Fusion for Place Recognition

We introduce a discriminative multimodal descriptor based on a pair of sensor readings: a point cloud from a LiDAR and an image from an RGB camera. Our descriptor, named MinkLoc++, can be used for place recognition, re-localization and loop closure purposes in robotics or autonomous vehicles applications. We use late fusion approach, where each modality is processed separately and fused in the final part of the processing pipeline. The proposed method achieves state-of-the-art performance on standard place recognition benchmarks. We also identify dominating modality problem when training a multimodal descriptor. The problem manifests itself when the network focuses on a modality with a larger overfit to the training data. This drives the loss down during the training but leads to suboptimal performance on the evaluation set. In this work we describe how to detect and mitigate such risk when using a deep metric learning approach to train a multimodal neural network. Our code is publicly available on the project website: https://github.com/jac99/MinkLocMultimodal.

cs.CV↗

arXiv2020

RegFlow: Probabilistic Flow-based Regression for Future Prediction

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 specific probability distributions that often poorly fit to real-life conditions. In this work we introduce a robust and flexible probabilistic framework that allows to model future predictions with virtually no constrains regarding the modality or underlying probability distribution. To achieve this goal, we leverage a hypernetwork architecture and train a continuous normalizing flow model. The resulting method dubbed RegFlow achieves state-of-the-art results on several benchmark datasets, outperforming competing approaches by a significant margin.

cs.LG↗

arXiv2020

Plugin Networks for Inference under Partial Evidence

In this paper, we propose a novel method to incorporate partial evidence in the inference of deep convolutional neural networks. Contrary to the existing, top performing methods, which either iteratively modify the input of the network or exploit external label taxonomy to take the partial evidence into account, we add separate network modules ("Plugin Networks") to the intermediate layers of a pre-trained convolutional network. The goal of these modules is to incorporate additional signal, ie information about known labels, into the inference procedure and adjust the predicted output accordingly. Since the attached plugins have a simple structure, consisting of only fully connected layers, we drastically reduced the computational cost of training and inference. At the same time, the proposed architecture allows to propagate information about known labels directly to the intermediate layers to improve the final representation. Extensive evaluation of the proposed method confirms that our Plugin Networks outperform the state-of-the-art in a variety of tasks, including scene categorization, multi-label image annotation, and semantic segmentation.

cs.CV↗

arXiv2019

SuperNCN: Neighbourhood consensus network for robust outdoor scenes matching

In this paper, we present a framework for computing dense keypoint correspondences between images under strong scene appearance changes. Traditional methods, based on nearest neighbour search in the feature descriptor space, perform poorly when environmental conditions vary, e.g. when images are taken at different times of the day or seasons. Our method improves finding keypoint correspondences in such difficult conditions. First, we use Neighbourhood Consensus Networks to build spatially consistent matching grid between two images at a coarse scale. Then, we apply Superpoint-like corner detector to achieve pixel-level accuracy. Both parts use features learned with domain adaptation to increase robustness against strong scene appearance variations. The framework has been tested on a RobotCar Seasons dataset, proving large improvement on pose estimation task under challenging environmental conditions.

cs.CV↗

arXiv2019

Late fusion of deep learning and hand-crafted features for Achilles tendon healing monitoring

Healing process assessment of the Achilles tendon is usually a complex procedure that relies on a combination of biomechanical and medical imaging tests. As a result, diagnostics remains a tedious and long-lasting task. Recently, a novel method for the automatic assessment of tendon healing based on Magnetic Resonance Imaging and deep learning was introduced. The method assesses six parameters related to the treatment progress utilizing a modified pre-trained network, PCA-reduced space, and linear regression. In this paper, we propose to improve this approach by incorporating hand-crafted features. We first perform a feature selection in order to obtain optimal sets of mixed hand-crafted and deep learning predictors. With the use of approx. 20,000 MRI slices, we then train a meta-regression algorithm that performs the tendon healing assessment. Finally, we evaluate the method against scores given by an experienced radiologist. In comparison with the previous baseline method, our approach significantly improves correlation in all of the six parameters assessed. Furthermore, our method uses only one MRI protocol and saves up to 60\% of the time needed for data acquisition.

eess.IV↗

arXiv2019

Monitoring Achilles tendon healing progress in ultrasound imaging with convolutional neural networks

Achilles tendon rupture is a debilitating injury, which is typically treated with surgical repair and long-term rehabilitation. The recovery, however, is protracted and often incomplete. Diagnosis, as well as healing progress assessment, are largely based on ultrasound and magnetic resonance imaging. In this paper, we propose an automatic method based on deep learning for analysis of Achilles tendon condition and estimation of its healing progress on ultrasound images. We develop custom convolutional neural networks for classification and regression on healing score and feature extraction. Our models are trained and validated on an acquired dataset of over 250.000 sagittal and over 450.000 axial ultrasound slices. The obtained estimates show a high correlation with the assessment of expert radiologists, with respect to all key parameters describing healing progress. We also observe that parameters associated with i.a. intratendinous healing processes are better modeled with sagittal slices. We prove that ultrasound imaging is quantitatively useful for clinical assessment of Achilles tendon healing process and should be viewed as complementary to magnetic resonance imaging.

eess.IV↗

arXiv2019

Random Binary Trees for Approximate Nearest Neighbour Search in Binary Space

Approximate nearest neighbour (ANN) search is one of the most important problems in computer science fields such as data mining or computer vision. In this paper, we focus on ANN for high-dimensional binary vectors and we propose a simple yet powerful search method that uses Random Binary Search Trees (RBST). We apply our method to a dataset of 1.25M binary local feature descriptors obtained from a real-life image-based localisation system provided by Google as a part of Project Tango. An extensive evaluation of our method against the state-of-the-art variations of Locality Sensitive Hashing (LSH), namely Uniform LSH and Multi-probe LSH, shows the superiority of our method in terms of retrieval precision with performance boost of over 20%

cs.CV↗

arXiv2019

DeepStyle: Multimodal Search Engine for Fashion and Interior Design

In this paper, we propose a multimodal search engine that combines visual and textual cues to retrieve items from a multimedia database aesthetically similar to the query. The goal of our engine is to enable intuitive retrieval of fashion merchandise such as clothes or furniture. Existing search engines treat textual input only as an additional source of information about the query image and do not correspond to the real-life scenario where the user looks for 'the same shirt but of denim'. Our novel method, dubbed DeepStyle, mitigates those shortcomings by using a joint neural network architecture to model contextual dependencies between features of different modalities. We prove the robustness of this approach on two different challenging datasets of fashion items and furniture where our DeepStyle engine outperforms baseline methods by 18-21% on the tested datasets. Our search engine is commercially deployed and available through a Web-based application.

cs.CV↗

arXiv2018

BinGAN: Learning Compact Binary Descriptors with a Regularized GAN

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-dimensional representation of the penultimate layer to mimic the distribution of the higher-dimensional preceding layers. To achieve this, we introduce two loss terms that aim at: (i) reducing the correlation between the dimensions of the binarized low-dimensional representation of the penultimate layer i. e. maximizing joint entropy) and (ii) propagating the relations between the dimensions in the high-dimensional space to the low-dimensional space. We evaluate the resulting binary image descriptors on two challenging applications, image matching and retrieval, and achieve state-of-the-art results.

cs.CV↗

arXiv2018

I Know How You Feel: Emotion Recognition with Facial Landmarks

Classification of human emotions remains an important and challenging task for many computer vision algorithms, especially in the era of humanoid robots which coexist with humans in their everyday life. Currently proposed methods for emotion recognition solve this task using multi-layered convolutional networks that do not explicitly infer any facial features in the classification phase. In this work, we postulate a fundamentally different approach to solve emotion recognition task that relies on incorporating facial landmarks as a part of the classification loss function. To that end, we extend a recently proposed Deep Alignment Network (DAN), that achieves state-of-the-art results in the recent facial landmark recognition challenge, with a term related to facial features. Thanks to this simple modification, our model called EmotionalDAN is able to outperform state-of-the-art emotion classification methods on two challenging benchmark dataset by up to 5%.

cs.CV↗