Uddeshya Upadhyay

Selected Publications

1. 

   BayesCap: Bayesian Identity Cap for Calibrated Uncertainty in Frozen Neural Networks

   Uddeshya U.*, Shyamgopal K.*, Yanbei C., Massimiliano M., Zeynep A.

   European Conference on Computer Vision (ECCV) 2022

   Abs Bib PDF

   High-quality calibrated uncertainty estimates are crucial for numerous real-world applications, especially for deep learning-based deployed ML systems. While Bayesian deep learning techniques allow uncertainty estimation, training them with large-scale datasets is an expensive process that does not always yield models competitive with non-Bayesian counterparts. Moreover, many of the high-performing deep learning models that are already trained and deployed are non-Bayesian in nature and do not provide uncertainty estimates. To address these issues, we propose BayesCap that learns a Bayesian identity mapping for the frozen model, allowing uncertainty estimation. BayesCap is a memory-efficient method that can be trained on a small fraction of the original dataset, enhancing pretrained non-Bayesian computer vision models by providing calibrated uncertainty estimates for the predictions without (i) hampering the performance of the model and (ii) the need for expensive retraining the model from scratch. The proposed method is agnostic to various architectures and tasks. We show the efficacy of our method on a wide variety of tasks with a diverse set of architectures, including image super-resolution, deblurring, inpainting, and crucial application such as medical image translation. Moreover, we apply the derived uncertainty estimates to detect out-of-distribution samples in critical scenarios like depth estimation in autonomous driving

   @article{uu_uncercyganneurips21,
     title = {BayesCap: Bayesian Identity Cap for Calibrated Uncertainty in Frozen Neural Networks},
     author = {{Uddeshya U.*}, {Shyamgopal K.*,} {Yanbei C.,} {Massimiliano M.,} {Zeynep A.}},
     year = {2022},
     journal = {European Conference on Computer Vision (ECCV)},
   }

2. 

   Robustness via Uncertainty-aware Cycle Consistency

   Uddeshya U., Yanbei C., Zeynep A.

   Advances in Neural Information Processing Systems (NeurIPS) 2021

   Abs Bib PDF

   Unpaired image-to-image translation refers to learning inter-image-domain mapping without corresponding image pairs. Existing methods learn deterministic mappings without explicitly modelling the robustness to outliers or predictive uncertainty, leading to performance degradation when encountering unseen perturbations at test time. To address this, we propose a novel probabilistic method based on Uncertainty-aware Generalized Adaptive Cycle Consistency (UGAC), which models the per-pixel residual by generalized Gaussian distribution, capable of modelling heavy-tailed distributions. We compare our model with a wide variety of state-of-the-art methods on various challenging tasks including unpaired image translation of natural images, using standard datasets, spanning autonomous driving, maps, facades, and also in medical imaging domain consisting of MRI. Experimental results demonstrate that our method exhibits stronger robustness towards unseen perturbations in test data.

   @article{uu_uncercyganneurips22,
     title = {Robustness via Uncertainty-aware Cycle Consistency},
     author = {{Uddeshya U.}, {Yanbei C.,} {Zeynep A.}},
     year = {2021},
     journal = {Advances in Neural Information Processing Systems (NeurIPS)},
   }

3. 

   Uncertainty-aware GAN with Adaptive Loss for Robust MRI Image Enhancement

   Uddeshya U., Viswanath P.S., Suyash P.A.

   ICCV workshop on CV for Automated Medical Diagnosis (ICCVw) 2021

   Abs Bib PDF

   Image-to-image translation is an ill-posed problem as unique one-to-one mapping may not exist between the source and target images. Learning-based methods proposed in this context often evaluate the performance on test data that is similar to the training data, which may be impractical. This demands robust methods that can quantify uncertainty in the prediction for making informed decisions, especially for critical areas such as medical imaging. Recent works that employ conditional generative adversarial networks (GANs) have shown improved performance in learning photo-realistic image-to-image mappings between the source and the target images. However, these methods do not focus on (i) robustness of the models to out-ofdistribution (OOD)-noisy data and (ii) uncertainty quantification. This paper proposes a GAN-based framework that (i) models an adaptive loss function for robustness to OOD-noisy data that automatically tunes the spatially varying norm for penalizing the residuals and (ii) estimates the per-voxel uncertainty in the predictions. We demonstrate our method on two key applications in medical imaging: (i) undersampled magnetic resonance imaging (MRI) reconstruction (ii) MRI modality propagation. Our experiments with two different real-world datasets show that the proposed method (i) is robust to OOD-noisy test data and provides improved accuracy and (ii) quantifies voxel-level uncertainty in the predictions.

   @article{uu_uncerganiccv21,
     title = {Uncertainty-aware GAN with Adaptive Loss for Robust MRI Image Enhancement},
     author = {{Uddeshya U.}, {Viswanath P.S.,} {Suyash P.A.}},
     year = {2021},
     journal = {ICCV workshop on CV for Automated Medical Diagnosis (ICCVw)},
   }

4. 

   Towards Lower-Dose PET using Physics-Based Uncertainty-Aware Multimodal Learning with Robustness to Out-of-Distribution Data (*joint first author)

   Viswanath P.S.*, Uddeshya U.*, Gary F.E., Zhaolin C., Suyash P.A.

   Medical Image Analysis (MedIA) 2021

   Abs Bib HTML PDF

   Radiation exposure in positron emission tomography (PET) imaging limits its usage in the studies of radiation-sensitive populations, e.g., pregnant women, children, and adults that require longitudinal imaging. Reducing the PET radiotracer dose or acquisition time reduces photon counts, which can deteriorate image quality. Recent deep-neural-network (DNN) based methods for image-to-image translation enable the mapping of low-quality PET images (acquired using substantially reduced dose), coupled with the associated magnetic resonance imaging (MRI) images, to high-quality PET images. However, such DNN methods focus on applications involving test data that match the statistical characteristics of the training data very closely and give little attention to evaluating the performance of these DNNs on new out-of-distribution (OOD) acquisitions. We propose a novel DNN formulation that models the (i) underlying sinogram-based physics of the PET imaging system and (ii) the uncertainty in the DNN output through the per-voxel heteroscedasticity of the residuals between the predicted and the high-quality reference images. Our sinogram-based uncertainty-aware DNN framework, namely, suDNN, estimates a standard-dose PET image using multimodal input in the form of (i) a low-dose/low-count PET image and (ii) the corresponding multi-contrast MRI images, leading to improved robustness of suDNN to OOD acquisitions. Results on in vivo simultaneous PET-MRI, and various forms of OOD data in PET-MRI, show the benefits of suDNN over the current state of the art, quantitatively and qualitatively.

   @article{uu_uncerphysmedia21,
     title = {Towards Lower-Dose PET using Physics-Based Uncertainty-Aware Multimodal Learning with Robustness to Out-of-Distribution Data (*joint first author)},
     author = {{Viswanath P.S.*}, {Uddeshya U.*,} {Gary F.E.,} {Zhaolin C.,} {Suyash P.A.}},
     year = {2021},
     journal = {Medical Image Analysis (MedIA)},
   }

5. 

   Uncertainty-Guided Progressive GANs for Medical Image Translation

   Uddeshya U., Yanbei C., Tobias H., Sergios G., Zeynep A.

   International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2021

   Abs Bib HTML PDF

   Image-to-image translation plays a vital role in tackling various medical imaging tasks such as attenuation correction, motion correction, undersampled reconstruction, and denoising. Generative adversarial networks have been shown to achieve the state-of-the-art in generating high fidelity images for these tasks. However, the state-of-the-art GANbased frameworks do not estimate the uncertainty in the predictions made by the network that is essential for making informed medical decisions and subsequent revision by medical experts and has recently been shown to improve the performance and interpretability of the model. In this work, we propose an uncertainty-guided progressive learning scheme for image-to-image translation. By incorporating aleatoric uncertainty as attention maps for GANs trained in a progressive manner, we generate images of increasing fidelity progressively. We demonstrate the efficacy of our model on three challenging medical image translation tasks, including PET to CT translation, undersampled MRI reconstruction, and MRI motion artefact correction. Our model generalizes well in three different tasks and improves performance over state of the art under fullsupervision and weak-supervision with limited data. Code is released here: https://github.com/ExplainableML/UncerGuidedI2I

   @article{uu_uncerguidedi2i,
     title = {Uncertainty-Guided Progressive GANs for Medical Image Translation},
     author = {{Uddeshya U.}, {Yanbei C.,} {Tobias H.,} {Sergios G.,} {Zeynep A.}},
     year = {2021},
     journal = {International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI)},
   }

6. 

   QUEST for MEDISYN: Quasi-norm based Uncertainty ESTimation for MEDical Image SYNthesis

   Uddeshya U., Viswanath P.S., Suyash P.A.

   ICML Workshop on Uncertainty & Robustness in Deep Learning (ICMLw) 2020

   Bib HTML

   @article{uu_icmlw1,
     title = {QUEST for MEDISYN: Quasi-norm based Uncertainty ESTimation for MEDical Image SYNthesis},
     author = {{Uddeshya U.}, {Viswanath P.S.,} {Suyash P.A.}},
     journal = {ICML Workshop on Uncertainty & Robustness in Deep Learning (ICMLw)},
     year = {2020},
   }

7. 

   A Mixed-Supervision Multilevel GAN Framework for Image Quality Enhancement

   Uddeshya U., Suyash P.A.

   International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2019

   Abs Bib HTML PDF

   Deep neural networks for image quality enhancement typically need large quantities of highly-curated training data comprising pairs of low-quality images and their corresponding high-quality images. While high-quality image acquisition is typically expensive and time-consuming, medium-quality images are faster to acquire, at lower equipment costs, and available in larger quantities. Thus, we propose a novel generative adversarial network (GAN) that can leverage training data at multiple levels of quality (e.g., high and medium quality) to improve performance while limiting costs of data curation. We apply our mixed-supervision GAN to (i) super-resolve histopathology images and (ii) enhance laparoscopy images by combining super-resolution and surgical smoke removal. Results on large clinical and pre-clinical datasets show the benefits of our mixed-supervision GAN over the state of the art.

   @article{uu_qegan,
     title = {A Mixed-Supervision Multilevel {GAN} Framework for Image Quality Enhancement},
     author = {{Uddeshya U.}, {Suyash P.A.}},
     pages = {556--564},
     year = {2019},
     doi = {10.1007/978-3-030-32254-0\_62},
     journal = {International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI)},
   }