Radhakrishna Achanta

À propos
Confidentialité
Mentions légales

Graph Chatbot

Publications associées (36)

TOPO-Loss for continuity-preserving crack detection using deep learning

Pascal Fua, Katrin Beyer, Radhakrishna Achanta, Bryan German Pantoja Rosero, Doruk Oner, Fernando Perez Cruz

We present a method for segmenting cracks in images of masonry buildings damaged by earthquakes. Existing methods of crack detection fail to preserve the continuity of cracks, and their performance deteriorates with imprecise training labels. We address th ...

2022

Machine-learning for damage assessment of rubble stone masonry piers based on crack patterns

Radhakrishna Achanta, Amir Rezaie

Under seismic actions, stone masonry buildings are prone to damage. To assess the severity of damaged masonry buildings and their failure modes, engineers connect these problems to surface crack features, such as the crack width and the extent of cracking. ...

ELSEVIER2022

Adaptive extreme video completion

Sabine Süsstrunk, Radhakrishna Achanta, Majed El Helou, Ruofan Zhou

A computer-implemented video completion method is proposed for reconstructing a sparsely sampled video comprising sparsely sampled video frames comprising sparse sets of picture element intensity values. The method comprises: distributing the sparse sets o ...

2021

Height And Weight Estimation From Unconstrained Images

Radhakrishna Achanta, Ntorina Thanou

We address the difficult problem of estimating the attributes of weight and height of individuals from pictures taken in completely unconstrained settings. We present a deep learning scheme that relies on simultaneous prediction of human silhouettes and sk ...

IEEE2020

Comparison of crack segmentation using digital image correlation measurements and deep learning

Katrin Beyer, Radhakrishna Achanta, Amir Rezaie, Michele Godio

Reliable methods for detecting pixels that represent cracks from laboratory images taken for digital image correlation (DIC) are required for two main reasons. Firstly, the segmented crack maps are used as an input for some DIC methods that are based on di ...

2020

Scale-Adaptive Superpixels

Pascal Fua, Sabine Süsstrunk, Radhakrishna Achanta, Pablo Marquez Neila

Size uniformity is one of the prominent features of superpixels. However, size uniformity rarely conforms to the varying content of an image. The chosen size of the superpixels therefore represents a compromise - how to obtain the fewest superpixels withou ...

Society for Imaging Science and Technology2018

Deep Feature Factorization for Concept Discovery

Sabine Süsstrunk, Radhakrishna Achanta, Edo Collins

We propose Deep Feature Factorization (DFF), a method capable of localizing similar semantic concepts within an image or a set of images. We use DFF to gain insight into a deep convolutional neural network's learned features, where we detect hierarchical c ...

SpringerLink2018

Deep Residual Network for Joint Demosaicing and Super- Resolution

Sabine Süsstrunk, Radhakrishna Achanta, Ruofan Zhou

The two classic image restoration tasks, demosaicing and super-resolution, have traditionally always been studied indepen- dently. That is sub-optimal as sequential processing, demosaic- ing and then super-resolution, may lead to amplification of ar- tifac ...

Society for Imaging Science and Technology ( IS&T )2018

Fourier-Domain Optimization for Image Processing

Sabine Süsstrunk, Radhakrishna Achanta, Majed El Helou, Frederike Dümbgen

Image optimization problems encompass many applications such as spectral fusion, deblurring, deconvolution, dehazing, matting, reflection removal and image interpolation, among others. With current image sizes in the order of megabytes, it is extremely exp ...

arXiv2018

Face Recognition in Real-world Images

Sabine Süsstrunk, Radhakrishna Achanta, Xavier Philippe Marie Fontaine

Face recognition systems are designed to handle well-aligned images captured under controlled situations. However real-world images present varying orientations, expressions, and illumination conditions. Traditional face recognition algorithms perform poor ...

Ieee2017

Page 2 sur 4