Vladimir A. Krylov

Evan Murphy, Marco Viola, Vladimir A. Krylov

In this paper we address the problem of precise geolocation of street furniture in complex urban environments, which is a critical task for effective monitoring and maintenance of public infrastructure by local authorities and private stakeholders. To this end, we propose a probabilistic framework based on energy maps that encode the spatial likelihood of object locations. Representing the energy in a map-based geopositioned format allows the optimisation process to seamlessly integrate external geospatial information, such as GIS layers, road maps, or placement constraints, which improves contextual awareness and localisation accuracy. A stochastic birth-and-death optimisation algorithm is introduced to infer the most probable configuration of assets. We evaluate our approach using a realistic simulation informed by a geolocated dataset of street lighting infrastructure in Dublin city centre, demonstrating its potential for scalable and accurate urban asset mapping. The implementation of the algorithm will be made available in the GitHub repository https://github.com/EMurphy0108/SBD_Street_Furniture.

Matej Ulicny, Vladimir A. Krylov, Julie Connelly et al.

We propose a pipeline for combined multi-class object geolocation and height estimation from street level RGB imagery, which is considered as a single available input data modality. Our solution is formulated via Markov Random Field optimization with deterministic output. The proposed technique uses image metadata along with coordinates of objects detected in the image plane as found by a custom-trained Convolutional Neural Network. Computing the object height using our methodology, in addition to object geolocation, has negligible effect on the overall computational cost. Accuracy is demonstrated experimentally for water drains and road signs on which we achieve average elevation estimation error lower than 20cm.

Chao-Jung Liu, Vladimir Krylov, Rozenn Dahyot

In this paper we propose an approach to perform semantic segmentation of 3D point cloud data by importing the geographic information from a 2D GIS layer (OpenStreetMap). The proposed automatic procedure identifies meaningful units such as buildings and adjusts their locations to achieve best fit between the GIS polygonal perimeters and the point cloud. Our processing pipeline is presented and illustrated by segmenting point cloud data of Trinity College Dublin (Ireland) campus constructed from optical imagery collected by a drone.

Matej Ulicny, Vladimir A. Krylov, Rozenn Dahyot

We show how parameter redundancy in Convolutional Neural Network (CNN) filters can be effectively reduced by pruning in spectral domain. Specifically, the representation extracted via Discrete Cosine Transform (DCT) is more conducive for pruning than the original space. By relying on a combination of weight tensor reshaping and reordering we achieve high levels of layer compression with just minor accuracy loss. Our approach is applied to compress pretrained CNNs and we show that minor additional fine-tuning allows our method to recover the original model performance after a significant parameter reduction. We validate our approach on ResNet-50 and MobileNet-V2 architectures for ImageNet classification task.

Matej Ulicny, Vladimir A. Krylov, Rozenn Dahyot

Convolutional neural networks (CNNs) learn filters in order to capture local correlation patterns in feature space. We propose to learn these filters as combinations of preset spectral filters defined by the Discrete Cosine Transform (DCT). Our proposed DCT-based harmonic blocks replace conventional convolutional layers to produce partially or fully harmonic versions of new or existing CNN architectures. Using DCT energy compaction properties, we demonstrate how the harmonic networks can be efficiently compressed by truncating high-frequency information in harmonic blocks thanks to the redundancies in the spectral domain. We report extensive experimental validation demonstrating benefits of the introduction of harmonic blocks into state-of-the-art CNN models in image classification, object detection and semantic segmentation applications.

Matej Ulicny, Vladimir A. Krylov, Rozenn Dahyot

Convolutional neural networks (CNNs) are very popular nowadays for image processing. CNNs allow one to learn optimal filters in a (mostly) supervised machine learning context. However this typically requires abundant labelled training data to estimate the filter parameters. Alternative strategies have been deployed for reducing the number of parameters and / or filters to be learned and thus decrease overfitting. In the context of reverting to preset filters, we propose here a computationally efficient harmonic block that uses Discrete Cosine Transform (DCT) filters in CNNs. In this work we examine the performance of harmonic networks in limited training data scenario. We validate experimentally that its performance compares well against scattering networks that use wavelets as preset filters.

Matej Ulicny, Vladimir A. Krylov, Rozenn Dahyot

Convolutional neural networks (CNNs) learn filters in order to capture local correlation patterns in feature space. In contrast, in this paper we propose harmonic blocks that produce features by learning optimal combinations of spectral filters defined by the Discrete Cosine Transform. The harmonic blocks are used to replace conventional convolutional layers to construct partial or fully harmonic CNNs. We extensively validate our approach and show that the introduction of harmonic blocks into state-of-the-art CNN baseline architectures results in comparable or better performance in classification tasks on small NORB, CIFAR10 and CIFAR100 datasets.