Search Results for author: Abhijit Guha Roy

Found 27 papers, 11 papers with code

Conformal prediction under ambiguous ground truth

no code implementations18 Jul 2023 David Stutz, Abhijit Guha Roy, Tatiana Matejovicova, Patricia Strachan, Ali Taylan Cemgil, Arnaud Doucet

However, in many real-world scenarios, the labels $Y_1,..., Y_n$ are obtained by aggregating expert opinions using a voting procedure, resulting in a one-hot distribution $\mathbb{P}_{vote}^{Y|X}$.

Conformal Prediction

A Simple Fix to Mahalanobis Distance for Improving Near-OOD Detection

3 code implementations16 Jun 2021 Jie Ren, Stanislav Fort, Jeremiah Liu, Abhijit Guha Roy, Shreyas Padhy, Balaji Lakshminarayanan

Mahalanobis distance (MD) is a simple and popular post-processing method for detecting out-of-distribution (OOD) inputs in neural networks.

Intent Detection Out-of-Distribution Detection +1

Importance Driven Continual Learning for Segmentation Across Domains

2 code implementations30 Apr 2020 Sinan Özgür Özgün, Anne-Marie Rickmann, Abhijit Guha Roy, Christian Wachinger

The ability of neural networks to continuously learn and adapt to new tasks while retaining prior knowledge is crucial for many applications.

Brain Segmentation Continual Learning +3

Recalibrating 3D ConvNets with Project & Excite

1 code implementation25 Feb 2020 Anne-Marie Rickmann, Abhijit Guha Roy, Ignacio Sarasua, Christian Wachinger

Fully Convolutional Neural Networks (F-CNNs) achieve state-of-the-art performance for segmentation tasks in computer vision and medical imaging.

Brain Segmentation Segmentation

`Project & Excite' Modules for Segmentation of Volumetric Medical Scans

2 code implementations11 Jun 2019 Anne-Marie Rickmann, Abhijit Guha Roy, Ignacio Sarasua, Nassir Navab, Christian Wachinger

Fully Convolutional Neural Networks (F-CNNs) achieve state-of-the-art performance for image segmentation in medical imaging.

Brain Segmentation Image Segmentation +2

BrainTorrent: A Peer-to-Peer Environment for Decentralized Federated Learning

no code implementations16 May 2019 Abhijit Guha Roy, Shayan Siddiqui, Sebastian Pölsterl, Nassir Navab, Christian Wachinger

A disadvantage of FL is the dependence on a central server, which requires all clients to agree on one trusted central body, and whose failure would disrupt the training process of all clients.

Brain Segmentation Federated Learning

Data Augmentation with Manifold Exploring Geometric Transformations for Increased Performance and Robustness

no code implementations14 Jan 2019 Magdalini Paschali, Walter Simson, Abhijit Guha Roy, Muhammad Ferjad Naeem, Rüdiger Göbl, Christian Wachinger, Nassir Navab

Compared with traditional augmentation methods, and with images synthesized by Generative Adversarial Networks our method not only achieves state-of-the-art performance but also significantly improves the network's robustness.

Data Augmentation General Classification +3

Bayesian QuickNAT: Model Uncertainty in Deep Whole-Brain Segmentation for Structure-wise Quality Control

2 code implementations24 Nov 2018 Abhijit Guha Roy, Sailesh Conjeti, Nassir Navab, Christian Wachinger

Next to voxel-wise uncertainty, we introduce four metrics to quantify structure-wise uncertainty in segmentation for quality control.

Brain Segmentation Segmentation +1

InfiNet: Fully Convolutional Networks for Infant Brain MRI Segmentation

no code implementations11 Oct 2018 Shubham Kumar, Sailesh Conjeti, Abhijit Guha Roy, Christian Wachinger, Nassir Navab

We present a novel, parameter-efficient and practical fully convolutional neural network architecture, termed InfiNet, aimed at voxel-wise semantic segmentation of infant brain MRI images at iso-intense stage, which can be easily extended for other segmentation tasks involving multi-modalities.

Infant Brain Mri Segmentation MRI segmentation +2

Recalibrating Fully Convolutional Networks with Spatial and Channel 'Squeeze & Excitation' Blocks

5 code implementations23 Aug 2018 Abhijit Guha Roy, Nassir Navab, Christian Wachinger

Towards this end, we introduce three variants of SE modules for segmentation, (i) squeezing spatially and exciting channel-wise, (ii) squeezing channel-wise and exciting spatially and (iii) joint spatial and channel 'squeeze & excitation'.

Image Classification Segmentation +1

SynNet: Structure-Preserving Fully Convolutional Networks for Medical Image Synthesis

no code implementations29 Jun 2018 Deepa Gunashekar, Sailesh Conjeti, Abhijit Guha Roy, Nassir Navab, Kuangyu Shi

Cross modal image syntheses is gaining significant interests for its ability to estimate target images of a different modality from a given set of source images, like estimating MR to MR, MR to CT, CT to PET etc, without the need for an actual acquisition. Though they show potential for applications in radiation therapy planning, image super resolution, atlas construction, image segmentation etc. The synthesis results are not as accurate as the actual acquisition. In this paper, we address the problem of multi modal image synthesis by proposing a fully convolutional deep learning architecture called the SynNet. We extend the proposed architecture for various input output configurations.

Image Generation Image Segmentation +2

Inherent Brain Segmentation Quality Control from Fully ConvNet Monte Carlo Sampling

no code implementations19 Apr 2018 Abhijit Guha Roy, Sailesh Conjeti, Nassir Navab, Christian Wachinger

We introduce inherent measures for effective quality control of brain segmentation based on a Bayesian fully convolutional neural network, using model uncertainty.

Brain Segmentation Segmentation +2

Concurrent Spatial and Channel Squeeze & Excitation in Fully Convolutional Networks

10 code implementations7 Mar 2018 Abhijit Guha Roy, Nassir Navab, Christian Wachinger

Fully convolutional neural networks (F-CNNs) have set the state-of-the-art in image segmentation for a plethora of applications.

Brain Segmentation Image Classification +4

QuickNAT: A Fully Convolutional Network for Quick and Accurate Segmentation of Neuroanatomy

6 code implementations12 Jan 2018 Abhijit Guha Roy, Sailesh Conjeti, Nassir Navab, Christian Wachinger

We introduce QuickNAT, a fully convolutional, densely connected neural network that segments a \revision{MRI brain scan} in 20 seconds.

Brain Segmentation Decision Making +2

Deep Residual Hashing

no code implementations16 Dec 2016 Sailesh Conjeti, Abhijit Guha Roy, Amin Katouzian, Nassir Navab

Hashing aims at generating highly compact similarity preserving code words which are well suited for large-scale image retrieval tasks.

Binarization Image Retrieval +3

Deep Neural Ensemble for Retinal Vessel Segmentation in Fundus Images towards Achieving Label-free Angiography

no code implementations19 Sep 2016 Avisek Lahiri, Abhijit Guha Roy, Debdoot Sheet, Prabir Kumar Biswas

Automated segmentation of retinal blood vessels in label-free fundus images entails a pivotal role in computed aided diagnosis of ophthalmic pathologies, viz., diabetic retinopathy, hypertensive disorders and cardiovascular diseases.

Retinal Vessel Segmentation Segmentation

DASA: Domain Adaptation in Stacked Autoencoders using Systematic Dropout

no code implementations19 Mar 2016 Abhijit Guha Roy, Debdoot Sheet

We experimentally evaluate performance in the problem of retinal vessel segmentation where the SAE-DNN is trained using large number of labeled samples in the source domain (DRIVE dataset) and adapted using less number of labeled samples in target domain (STARE dataset).

Domain Adaptation Retinal Vessel Segmentation

Cannot find the paper you are looking for? You can Submit a new open access paper.