no code implementations • ECCV 2020 • Poojan Oza, Hien V. Nguyen, Vishal M. Patel
To this end, we consider the problem of multiple class novelty detection under dataset distribution shift to improve the novelty detection performance.
no code implementations • ECCV 2020 • Poojan Oza, Vishal M. Patel
For any recognition system, the ability to identify novel class samples during inference is an important aspect of the system’s robustness.
1 code implementation • 17 Jan 2025 • Kartik Narayan, Vibashan VS, Vishal M. Patel
To address this gap, we introduce FaceXBench, a comprehensive benchmark designed to evaluate MLLMs on complex face understanding tasks.
no code implementations • 16 Jan 2025 • Deepti Hegde, Rajeev Yasarla, Hong Cai, Shizhong Han, Apratim Bhattacharyya, Shweta Mahajan, Litian Liu, Risheek Garrepalli, Vishal M. Patel, Fatih Porikli
Training with DiMA results in a 37% reduction in the L2 trajectory error and an 80% reduction in the collision rate of the vision-based planner, as well as a 44% trajectory error reduction in longtail scenarios.
no code implementations • 13 Jan 2025 • Yasiru Ranasinghe, Vibashan VS, James Uplinger, Celso de Melo, Vishal M. Patel
In extreme use cases, such as military applications, these factors are often challenged due to the presence of unknown terrains, environmental conditions, and novel object categories.
2 code implementations • 11 Dec 2024 • Kartik Narayan, Vibashan VS, Vishal M. Patel
Face parsing refers to the semantic segmentation of human faces into key facial regions such as eyes, nose, hair, etc.
Ranked #2 on
Face Parsing
on CelebAMask-HQ
no code implementations • 10 Dec 2024 • Kartik Narayan, Nithin Gopalakrishnan Nair, Jennifer Xu, Rama Chellappa, Vishal M. Patel
(1) We solve catastrophic forgetting by leveraging the power of parameter efficient fine-tuning(PEFT).
no code implementations • 26 Nov 2024 • Sudarshan Rajagopalan, Vishal M. Patel
AWIR models are trained to address a specific set of weather conditions such as fog, rain, and snow.
no code implementations • 26 Nov 2024 • Sudarshan Rajagopalan, Nithin Gopalakrishnan Nair, Jay N. Paranjape, Vishal M. Patel
Using GenDeg, we synthesize over 550k samples across six degradation types: haze, rain, snow, motion blur, low-light, and raindrops.
1 code implementation • 31 Oct 2024 • Jay N. Paranjape, Shameema Sikder, S. Swaroop Vedula, Vishal M. Patel
Federated Learning (FL) is a form of distributed learning that allows multiple institutions or clients to collaboratively learn a global model to solve a task.
no code implementations • 29 Oct 2024 • Bardia Safaei, Vishal M. Patel
Pre-trained vision-language models (VLMs) like CLIP have demonstrated impressive zero-shot performance on a wide range of downstream computer vision tasks.
1 code implementation • 19 Sep 2024 • Yilmaz Korkmaz, Vishal M. Patel
Unpaired image-to-image translation is a challenging task due to the absence of paired examples, which complicates learning the complex mappings between the distinct distributions of the source and target domains.
1 code implementation • 19 Sep 2024 • Yilmaz Korkmaz, Vishal M. Patel
Magnetic Resonance Imaging (MRI) is one of the most important medical imaging modalities as it provides superior resolution of soft tissues, albeit with a notable limitation in scanning speed.
no code implementations • 30 Aug 2024 • Sudarshan Rajagopalan, Vishal M. Patel
In this paper, we propose All-Weather Image Restoration using Visual In-Context Learning (AWRaCLe), a novel approach for AWIR that innovatively utilizes degradation-specific visual context information to steer the image restoration process.
1 code implementation • 12 Aug 2024 • Jay N. Paranjape, Shameema Sikder, S. Swaroop Vedula, Vishal M. Patel
Our experiments show that S-SAM outperforms state-of-the-art methods as well as existing SAM adaptation methods while tuning a significantly less number of parameters.
no code implementations • 13 Jul 2024 • Ruihuang Li, Zhengqiang Zhang, Chenhang He, Zhiyuan Ma, Vishal M. Patel, Lei Zhang
Recent vision-language pre-training models have exhibited remarkable generalization ability in zero-shot recognition tasks.
no code implementations • 9 Jul 2024 • Yiqun Mei, Jiacong Xu, Vishal M. Patel
Simply optimizing the appearance as prior methods do is often insufficient for modeling continuous textures in the given reference image.
1 code implementation • 8 Jul 2024 • Jay N. Paranjape, Celso de Melo, Vishal M. Patel
Change detection in remote sensing images is an essential tool for analyzing a region at different times.
no code implementations • CVPR 2024 • Yu Zeng, Vishal M. Patel, Haochen Wang, Xun Huang, Ting-Chun Wang, Ming-Yu Liu, Yogesh Balaji
Personalized text-to-image generation models enable users to create images that depict their individual possessions in diverse scenes, finding applications in various domains.
no code implementations • 25 Jun 2024 • Ruihuang Li, Liyi Chen, Zhengqiang Zhang, Varun Jampani, Vishal M. Patel, Lei Zhang
Meanwhile, the 2D diffusion models also exhibit substantial potentials for 3D editing tasks.
no code implementations • 19 Jun 2024 • Ke Zhang, Vishal M. Patel
Pixel-level dense labeling is both resource-intensive and time-consuming, whereas weak labels such as scribble present a more feasible alternative to full annotations.
no code implementations • 14 Jun 2024 • Jiacong Xu, Yiqun Mei, Vishal M. Patel
Unlike previous methods that model reference features in image space, Wild-GS explicitly aligns the pixel appearance features to the corresponding local Gaussians by sampling the triplane extracted from the reference image.
no code implementations • 20 May 2024 • Shao-Yuan Lo, Vishal M. Patel
In this paper, we aim at a non-AT defense: How to design a defense method that gets rid of AT but is still robust against strong adversarial attacks?
1 code implementation • 17 May 2024 • Jay N. Paranjape, Shameema Sikder, S. Swaroop Vedula, Vishal M. Patel
There have been some attempts in the literature to perform parameter-efficient finetuning of such foundation models for medical image segmentation.
no code implementations • 22 Apr 2024 • Kartik Narayan, Vishal M. Patel
Most prior research in face anti-spoofing (FAS) approaches it as a two-class classification task where models are trained on real samples and known spoof attacks and tested for detection performance on unknown spoof attacks.
1 code implementation • 18 Apr 2024 • Sina Sharifi, Taha Entesari, Bardia Safaei, Vishal M. Patel, Mahyar Fazlyab
In this work, we propose the idea of leveraging the information embedded in the gradient of the loss function during training to enable the network to not only learn a desired OOD score for each sample but also to exhibit similar behavior in a local neighborhood around each sample.
Out-of-Distribution Detection
Out of Distribution (OOD) Detection
no code implementations • 17 Apr 2024 • Deepti Hegde, Suhas Lohit, Kuan-Chuan Peng, Michael J. Jones, Vishal M. Patel
This can enable improved performance in downstream tasks that are equivariant to such transformations.
no code implementations • 17 Apr 2024 • Deepti Hegde, Suhas Lohit, Kuan-Chuan Peng, Michael J. Jones, Vishal M. Patel
To this end, we propose CLIX$^\text{3D}$, a multimodal fusion and supervised contrastive learning framework for 3D object detection that performs alignment of object features from same-class samples of different domains while pushing the features from different classes apart.
no code implementations • 15 Apr 2024 • Nithin Gopalakrishnan Nair, Jeya Maria Jose Valanarasu, Vishal M. Patel
As these parameters are independent, a single diffusion model with these task-specific parameters can be used to perform multiple tasks simultaneously.
no code implementations • 1 Apr 2024 • Kangfu Mei, Zhengzhong Tu, Mauricio Delbracio, Hossein Talebi, Vishal M. Patel, Peyman Milanfar
We study the scaling properties of latent diffusion models (LDMs) with an emphasis on their sampling efficiency.
no code implementations • 28 Mar 2024 • Aimon Rahman, Malsha V. Perera, Vishal M. Patel
In our paper, we present a systematic investigation into the phenomenon of sample replication in video diffusion models.
1 code implementation • 21 Mar 2024 • Jiacong Xu, Mingqian Liao, K Ram Prabhakar, Vishal M. Patel
To address these issues, we present Thermal-NeRF, which takes thermal and visible raw images as inputs, considering the thermal camera is robust to the illumination variation and raw images preserve any possible clues in the dark, to accomplish visible and thermal view synthesis simultaneously.
1 code implementation • CVPR 2024 • Quan Zhang, Lei Wang, Vishal M. Patel, Xiaohua Xie, JianHuang Lai
Experiments on two datasets show that VDT is a feasible and effective solution for AGPReID, surpassing the previous method on mAP/Rank1 by up to 5. 0%/2. 7% on CARGO and 3. 7%/5. 2% on AG-ReID, keeping the same magnitude of computational complexity.
Ranked #1 on
Person Re-Identification
on AG-ReID
2 code implementations • 19 Mar 2024 • Kartik Narayan, Vibashan VS, Rama Chellappa, Vishal M. Patel
In this work, we introduce FaceXFormer, an end-to-end unified transformer model capable of performing nine facial analysis tasks including face parsing, landmark detection, head pose estimation, attribute prediction, and estimation of age, gender, race, expression, and face visibility within a single framework.
no code implementations • CVPR 2024 • Yiqun Mei, Yu Zeng, He Zhang, Zhixin Shu, Xuaner Zhang, Sai Bi, Jianming Zhang, HyunJoon Jung, Vishal M. Patel
At the core of portrait photography is the search for ideal lighting and viewpoint.
1 code implementation • 11 Mar 2024 • Wele Gedara Chaminda Bandara, Vishal M. Patel
This approach greatly reduces the number of learnable parameters compared to full tuning.
no code implementations • 27 Feb 2024 • Mo Zhou, Yiding Yang, Haoxiang Li, Vishal M. Patel, Gang Hua
With a strong alignment between the training and test distributions, object relation as a context prior facilitates object detection.
1 code implementation • 3 Feb 2024 • Yatong Bai, Mo Zhou, Vishal M. Patel, Somayeh Sojoudi
Adversarial robustness often comes at the cost of degraded accuracy, impeding real-life applications of robust classification models.
no code implementations • CVPR 2024 • Yasiru Ranasinghe, Deepti Hegde, Vishal M. Patel
Hence we adopt a Gaussian mixture model to sample noise during the forward diffusion process and initialize the reverse diffusion process.
no code implementations • CVPR 2024 • Yasiru Ranasinghe, Nithin Gopalakrishnan Nair, Wele Gedara Chaminda Bandara, Vishal M. Patel
In addition owing to the stochastic nature of the diffusion model we introduce producing multiple density maps to improve the counting performance contrary to the existing crowd counting pipelines.
no code implementations • CVPR 2024 • Nisarg A. Shah, Vibashan VS, Vishal M. Patel
To address this issue we propose a Multi-modal Query Feature Fusion technique characterized by two innovative designs: (1) Gaussian enhanced Multi-Modal Fusion a novel visual grounding mechanism that enhances overall representation by extracting rich local visual information and global visual-linguistic relationships and (2) A Dynamic Query Module that produces a diverse set of queries through a scoring network where the network selectively focuses on queries for objects referred to in the language description.
2 code implementations • 21 Dec 2023 • Bardia Safaei, Vibashan VS, Celso M. de Melo, Vishal M. Patel
Active Learning (AL) aims to enhance the performance of deep models by selecting the most informative samples for annotation from a pool of unlabeled data.
no code implementations • 4 Dec 2023 • Kangfu Mei, Luis Figueroa, Zhe Lin, Zhihong Ding, Scott Cohen, Vishal M. Patel
Recovering textures under shadows has remained a challenging problem due to the difficulty of inferring shadow-free scenes from shadow images.
1 code implementation • 4 Dec 2023 • Wele Gedara Chaminda Bandara, Celso M. de Melo, Vishal M. Patel
Self-supervised Learning (SSL) aims to learn transferable feature representations for downstream applications without relying on labeled data.
Ranked #1 on
Self-Supervised Learning
on STL-10
1 code implementation • CVPR 2024 • Kangfu Mei, Mauricio Delbracio, Hossein Talebi, Zhengzhong Tu, Vishal M. Patel, Peyman Milanfar
Our conditional-task learning and distillation approach outperforms previous distillation methods, achieving a new state-of-the-art in producing high-quality images with very few steps (e. g., 1-4) across multiple tasks, including super-resolution, text-guided image editing, and depth-to-image generation.
1 code implementation • ICCV 2023 • Nithin Gopalakrishnan Nair, Anoop Cherian, Suhas Lohit, Ye Wang, Toshiaki Koike-Akino, Vishal M. Patel, Tim K. Marks
To this end, and capitalizing on the powerful fine-grained generative control offered by the recent diffusion-based generative models, we introduce Steered Diffusion, a generalized framework for photorealistic zero-shot conditional image generation using a diffusion model trained for unconditional generation.
no code implementations • 11 Sep 2023 • Pengfei Guo, Warren Richard Morningstar, Raviteja Vemulapalli, Karan Singhal, Vishal M. Patel, Philip Andrew Mansfield
To mitigate this issue and facilitate training of large models on edge devices, we introduce a simple yet effective strategy, Federated Layer-wise Learning, to simultaneously reduce per-client memory, computation, and communication costs.
1 code implementation • 7 Aug 2023 • Jay N. Paranjape, Nithin Gopalakrishnan Nair, Shameema Sikder, S. Swaroop Vedula, Vishal M. Patel
However, SAM does not generalize well to the medical domain as is without utilizing a large amount of compute resources for fine-tuning and using task-specific prompts.
1 code implementation • 31 Jul 2023 • Jeya Maria Jose Valanarasu, Yucheng Tang, Dong Yang, Ziyue Xu, Can Zhao, Wenqi Li, Vishal M. Patel, Bennett Landman, Daguang Xu, Yufan He, Vishwesh Nath
We curate a large-scale dataset to enable pre-training of 3D medical radiology images (MRI and CT).
1 code implementation • 31 Jul 2023 • Jay N. Paranjape, Shameema Sikder, Vishal M. Patel, S. Swaroop Vedula
In this paper, we highlight this domain shift in the commonly performed cataract surgery and propose a novel end-to-end Unsupervised Domain Adaptation (UDA) method called the Barlow Adaptor that addresses the problem of distribution shift without requiring any labels from another domain.
1 code implementation • 20 Jul 2023 • Nisarg A. Shah, Shameema Sikder, S. Swaroop Vedula, Vishal M. Patel
These results validate the suitability of our proposed approach for automated surgical step recognition.
1 code implementation • 29 Jun 2023 • Yilmaz Korkmaz, Tolga Cukur, Vishal M. Patel
Magnetic Resonance Imaging (MRI) produces excellent soft tissue contrast, albeit it is an inherently slow imaging modality.
1 code implementation • 25 May 2023 • Yu Zeng, Mo Zhou, Yuan Xue, Vishal M. Patel
Prior research attempted to mitigate these threats by detecting generated images, but the varying traces left by different generative models make it challenging to create a universal detector capable of generalizing to new, unseen generative models.
no code implementations • 24 May 2023 • Kangfu Mei, Mo Zhou, Vishal M. Patel
The model can be scaled to generate high-resolution data while unifying multiple modalities.
no code implementations • 10 May 2023 • Malsha V. Perera, Vishal M. Patel
Diffusion models are becoming increasingly popular in synthetic data generation and image editing applications.
no code implementations • CVPR 2023 • Vibashan VS, Ning Yu, Chen Xing, Can Qin, Mingfei Gao, Juan Carlos Niebles, Vishal M. Patel, ran Xu
In summary, an OV method learns task-specific information using strong supervision from base annotations and novel category information using weak supervision from image-captions pairs.
1 code implementation • CVPR 2023 • Shao-Yuan Lo, Poojan Oza, Sumanth Chennupati, Alejandro Galindo, Vishal M. Patel
Unsupervised Domain Adaptation (UDA) of semantic segmentation transfers labeled source knowledge to an unlabeled target domain by relying on accessing both the source and target data.
Ranked #1 on
Source-Free Domain Adaptation
on VIPER-to-Cityscapes
no code implementations • 23 Mar 2023 • Jeya Maria Jose Valanarasu, Rahul Garg, Andeep Toor, Xin Tong, Weijuan Xi, Andreas Lugmayr, Vishal M. Patel, Anne Menini
The first branch learns spatio-temporal features by tokenizing the input frames along the spatial and temporal dimensions using a ConvNext-based encoder and processing these abstract tokens using a bottleneck mixer.
no code implementations • CVPR 2023 • Yiqun Mei, He Zhang, Xuaner Zhang, Jianming Zhang, Zhixin Shu, Yilin Wang, Zijun Wei, Shi Yan, HyunJoon Jung, Vishal M. Patel
Recent portrait relighting methods have achieved realistic results of portrait lighting effects given a desired lighting representation such as an environment map.
1 code implementation • 22 Mar 2023 • Yasiru Ranasinghe, Nithin Gopalakrishnan Nair, Wele Gedara Chaminda Bandara, Vishal M. Patel
Furthermore, as the intermediate time steps of the diffusion process are noisy, we incorporate a regression branch for direct crowd estimation only during training to improve the feature learning.
1 code implementation • 20 Mar 2023 • Deepti Hegde, Jeya Maria Jose Valanarasu, Vishal M. Patel
Attempting to train the visual and text encoder to account for this shift results in catastrophic forgetting and a notable decrease in performance.
1 code implementation • 16 Mar 2023 • Wele Gedara Chaminda Bandara, Vishal M. Patel
This loss is motivated by the principle of metric learning where we simultaneously maximize the distance between change pair-wise pixels while minimizing the distance between no-change pair-wise pixels in bi-temporal image domain and their deep feature domain.
no code implementations • 15 Mar 2023 • Huali Xu, Shuaifeng Zhi, Shuzhou Sun, Vishal M. Patel, Li Liu
To address this, Few-shot learning (FSL) enables models to perform the target tasks with very few labeled examples by leveraging prior knowledge from related tasks.
no code implementations • 14 Dec 2022 • Kangfu Mei, Nithin Gopalakrishnan Nair, Vishal M. Patel
The improvements obtained by our method suggest that the priors can be incorporated as a general plugin for improving conditional diffusion models.
1 code implementation • CVPR 2023 • Nithin Gopalakrishnan Nair, Wele Gedara Chaminda Bandara, Vishal M. Patel
We also introduce a novel reliability parameter that allows using different off-the-shelf diffusion models trained across various datasets during sampling time alone to guide it to the desired outcome satisfying multiple constraints.
Ranked #1 on
Face Sketch Synthesis
on Multi-Modal CelebA-HQ
1 code implementation • 1 Dec 2022 • Kangfu Mei, Vishal M. Patel
Diffusion models have emerged as a powerful generative method for synthesizing high-quality and diverse set of images.
Ranked #18 on
Video Generation
on UCF-101
no code implementations • CVPR 2023 • Yu Zeng, Zhe Lin, Jianming Zhang, Qing Liu, John Collomosse, Jason Kuen, Vishal M. Patel
We propose a new framework for conditional image synthesis from semantic layouts of any precision levels, ranging from pure text to a 2D semantic canvas with precise shapes.
2 code implementations • CVPR 2023 • Wele Gedara Chaminda Bandara, Naman Patel, Ali Gholami, Mehdi Nikkhah, Motilal Agrawal, Vishal M. Patel
Our adaptive masking strategy samples visible tokens based on the semantic context using an auxiliary sampling network.
Ranked #1 on
Action Classification
on Something-Something V2
1 code implementation • 10 Nov 2022 • Bardia Safaei, Vibashan VS, Celso M. de Melo, Shuowen Hu, Vishal M. Patel
Automatic Target Recognition (ATR) is a category of computer vision algorithms which attempts to recognize targets on data obtained from different sensors.
no code implementations • 20 Sep 2022 • Nithin Gopalakrishnan Nair, Rajeev Yasarla, Vishal M. Patel
This results in a pair of images with colored noise.
1 code implementation • 19 Sep 2022 • Nithin Gopalakrishnan Nair, Vishal M. Patel
In this paper, we propose a Denoising Diffusion Probabilistic Model (DDPM) based solution for T2V translation specifically for facial images.
1 code implementation • 24 Aug 2022 • Nithin Gopalakrishnan Nair, Kangfu Mei, Vishal M. Patel
In recent years, various deep learning-based single image atmospheric turbulence mitigation methods, including CNN-based and GAN inversion-based, have been proposed in the literature which attempt to remove the distortion in the image.
no code implementations • 30 Jul 2022 • Shao-Yuan Lo, Wei Wang, Jim Thomas, Jingjing Zheng, Vishal M. Patel, Cheng-Hao Kuo
In this paper, we propose a novel UDA method for MDE, referred to as Learning Feature Decomposition for Adaptation (LFDA), which learns to decompose the feature space into content and style components.
1 code implementation • 19 Jul 2022 • Kangfu Mei, Vishal M. Patel, Rui Huang
The ultimate aim of image restoration like denoising is to find an exact correlation between the noisy and clear image domains.
1 code implementation • 7 Jul 2022 • Rajeev Yasarla, Vishal M. Patel
Atmospheric turbulence can significantly degrade the quality of images acquired by long-range imaging systems by causing spatially and temporally random fluctuations in the index of refraction of the atmosphere.
1 code implementation • 23 Jun 2022 • Wele Gedara Chaminda Bandara, Nithin Gopalakrishnan Nair, Vishal M. Patel
However, in this work, our focus is not on image synthesis but on utilizing it as a pre-trained feature extractor for the downstream application of change detection.
Ranked #1 on
Change Detection
on DSIFN-CD
1 code implementation • 9 Jun 2022 • Malsha V. Perera, Nithin Gopalakrishnan Nair, Wele Gedara Chaminda Bandara, Vishal M. Patel
The despeckled image is recovered by a reverse process which iteratively predicts the added noise using a noise predictor which is conditioned on the speckled image.
1 code implementation • 31 May 2022 • Malsha V. Perera, Wele Gedara Chaminda Bandara, Jeya Maria Jose Valanarasu, Vishal M. Patel
We show that the proposed network improves despeckling performance compared to recent despeckling methods on synthetic and real SAR images.
no code implementations • 19 May 2022 • Mo Zhou, Vishal M. Patel
Adversarial attacks pose safety and security concerns to deep learning applications, but their characteristics are under-explored.
1 code implementation • 25 Apr 2022 • Xirui Hou, Pengfei Guo, Puyang Wang, Peiying Liu, Doris D. M. Lin, Hongli Fan, Yang Li, Zhiliang Wei, Zixuan Lin, Dengrong Jiang, Jin Jin, Catherine Kelly, Jay J. Pillai, Judy Huang, Marco C. Pinho, Binu P. Thomas, Babu G. Welch, Denise C. Park, Vishal M. Patel, Argye E. Hillis, Hanzhang Lu
Deep-learning resting-state vascular imaging has the potential to become a useful tool in clinical cerebrovascular imaging.
no code implementations • 23 Apr 2022 • Rajeev Yasarla, Vishwanath A. Sindagi, Vishal M. Patel
Existing approaches for restoring weather-degraded images follow a fully-supervised paradigm and they require paired data for training.
no code implementations • 19 Apr 2022 • Nithin Gopalakrishnan Nair, Kangfu Mei, Vishal M. Patel
In this paper, we systematically evaluate the effectiveness of various turbulence simulation methods on image restoration.
1 code implementation • 18 Apr 2022 • Wele Gedara Chaminda Bandara, Vishal M. Patel
The performance of existing deep supervised CD methods is attributed to the large amounts of annotated data used to train the networks.
no code implementations • 16 Apr 2022 • Yu Zeng, Zhe Lin, Vishal M. Patel
Therefore, we propose a new data preparation method and a novel Contextual Object Generator (CogNet) for the object inpainting task.
2 code implementations • 11 Apr 2022 • Vibashan VS, Poojan Oza, Vishal M. Patel
To the best of our knowledge, this is the first work to address online and offline adaptation settings for object detection.
no code implementations • 6 Apr 2022 • Kangfu Mei, Yiqun Mei, Vishal M. Patel
In this paper, we first investigate the problem with a turbulence simulation method on real-world thermal images.
no code implementations • CVPR 2022 • Yiqun Mei, Pengfei Guo, Vishal M. Patel
In Heterogeneous Face Recognition (HFR), the objective is to match faces across two different domains such as visible and thermal.
1 code implementation • CVPR 2023 • Vibashan VS, Poojan Oza, Vishal M. Patel
The Source-Free Domain Adaptation (SFDA) setting aims to alleviate these concerns by adapting a source-trained model for the target domain without requiring access to the source data.
1 code implementation • 29 Mar 2022 • Vibashan VS, Jeya Maria Jose Valanarasu, Vishal M. Patel
In task-specific adaptation, we exploit the enhanced pseudo-labels using a student-teacher framework to effectively learn segmentation on the target domain.
no code implementations • 15 Mar 2022 • Jeya Maria Jose Valanarasu, He Zhang, Jianming Zhang, Yilin Wang, Zhe Lin, Jose Echevarria, Yinglan Ma, Zijun Wei, Kalyan Sunkavalli, Vishal M. Patel
To enable flexible interaction between user and harmonization, we introduce interactive harmonization, a new setting where the harmonization is performed with respect to a selected \emph{region} in the reference image instead of the entire background.
no code implementations • 12 Mar 2022 • Pengfei Guo, Dong Yang, Ali Hatamizadeh, An Xu, Ziyue Xu, Wenqi Li, Can Zhao, Daguang Xu, Stephanie Harmon, Evrim Turkbey, Baris Turkbey, Bradford Wood, Francesca Patella, Elvira Stellato, Gianpaolo Carrafiello, Vishal M. Patel, Holger R. Roth
Federated learning (FL) is a distributed machine learning technique that enables collaborative model training while avoiding explicit data sharing.
1 code implementation • 10 Mar 2022 • Jeya Maria Jose Valanarasu, Pengfei Guo, Vibashan VS, Vishal M. Patel
During test-time, the model takes in just the new test image and generates a domain code to adapt the features of source model according to the test data.
2 code implementations • 9 Mar 2022 • Jeya Maria Jose Valanarasu, Vishal M. Patel
Using tokenized MLPs in latent space reduces the number of parameters and computational complexity while being able to result in a better representation to help segmentation.
Ranked #4 on
Medical Image Segmentation
on ISIC 2018
1 code implementation • CVPR 2022 • Wele Gedara Chaminda Bandara, Vishal M. Patel
Existing pansharpening approaches neglect using an attention mechanism to transfer HR texture features from PAN to LR-HSI features, resulting in spatial and spectral distortions.
2 code implementations • CVPR 2022 • Mo Zhou, Vishal M. Patel
Owing to security implications of adversarial vulnerability, adversarial robustness of deep metric learning models has to be improved.
1 code implementation • 18 Feb 2022 • Shao-Yuan Lo, Vishal M. Patel
Adversarial Training (AT) has been considered to be the most successful adversarial defense approach.
1 code implementation • 12 Feb 2022 • Rui Shao, Pramuditha Perera, Pong C. Yuen, Vishal M. Patel
This paper proposes an Open-Set Defense Network with Clean-Adversarial Mutual Learning (OSDN-CAML) as a solution to the OSAD problem.
1 code implementation • 23 Jan 2022 • Pengfei Guo, Yiqun Mei, Jinyuan Zhou, Shanshan Jiang, Vishal M. Patel
Accelerating magnetic resonance image (MRI) reconstruction process is a challenging ill-posed inverse problem due to the excessive under-sampling operation in k-space.
1 code implementation • 23 Jan 2022 • Malsha V. Perera, Wele Gedara Chaminda Bandara, Jeya Maria Jose Valanarasu, Vishal M. Patel
Synthetic Aperture Radar (SAR) images are usually degraded by a multiplicative noise known as speckle which makes processing and interpretation of SAR images difficult.
3 code implementations • 4 Jan 2022 • Wele Gedara Chaminda Bandara, Vishal M. Patel
This paper presents a transformer-based Siamese network architecture (abbreviated by ChangeFormer) for Change Detection (CD) from a pair of co-registered remote sensing images.
Ranked #21 on
Change Detection
on LEVIR-CD
no code implementations • 4 Dec 2021 • Kangfu Mei, Vishal M. Patel
To mitigate the turbulence effect, in this paper, we propose the first turbulence mitigation method that makes use of visual priors encapsulated by a well-trained GAN.
no code implementations • CVPR 2022 • Yu Zeng, Zhe Lin, Vishal M. Patel
Our model can be trained in a self-supervised fashion by learning the reconstruction of an image region from the style vector and sketch.
1 code implementation • 30 Nov 2021 • Deepti Hegde, Vishal M. Patel
We demonstrate our approach on two recent object detectors and achieve results that out-perform the other domain adaptation works.
1 code implementation • CVPR 2022 • Jeya Maria Jose Valanarasu, Rajeev Yasarla, Vishal M. Patel
We also introduce a transformer decoder with learnable weather type embeddings to adjust to the weather degradation at hand.
Ranked #1 on
Single Image Deraining
on Raindrop
no code implementations • 18 Nov 2021 • Pengfei Guo, Vishal M. Patel
Deep Learning (DL) based methods for magnetic resonance (MR) image reconstruction have been shown to produce superior performance in recent years.
no code implementations • 25 Oct 2021 • Rui Shao, Bochao Zhang, Pong C. Yuen, Vishal M. Patel
The generalization ability of face presentation attack detection models to unseen attacks has become a key issue for real-world deployment, which can be improved when models are trained with face images from different input distributions and different types of spoof attacks.
no code implementations • 21 Oct 2021 • Shraman Pramanick, Aniket Roy, Vishal M. Patel
Multimodal learning is an emerging yet challenging research area.
no code implementations • 7 Oct 2021 • Vibashan VS, Domenick Poster, Suya You, Shuowen Hu, Vishal M. Patel
Though thermal cameras are widely used for military applications and increasingly for commercial applications, there is a lack of robust algorithms to robustly exploit the thermal imagery due to the limited availability of labeled thermal data.
no code implementations • 20 Sep 2021 • Jeya Maria Jose Valanarasu, Vishal M. Patel
First, we propose a Fine Context-aware Shadow Detection Network (FCSD-Net), where we constraint the receptive field size and focus on low-level features to learn fine context features better.
1 code implementation • 16 Sep 2021 • Wele Gedara Chaminda Bandara, Jeya Maria Jose Valanarasu, Vishal M. Patel
Using just convolution neural networks (ConvNets) for this problem is not effective as it is inefficient at capturing distant dependencies between road segments in the image which is essential to extract road connectivity.
Ranked #1 on
Road Segmentation
on DeepGlobe
1 code implementation • 25 Aug 2021 • Shao-Yuan Lo, Poojan Oza, Vishal M. Patel
To this end, we propose a defense strategy that manipulates the latent space of novelty detectors to improve the robustness against adversarial examples.
no code implementations • 21 Aug 2021 • Neehar Peri, Joshua Gleason, Carlos D. Castillo, Thirimachos Bourlai, Vishal M. Patel, Rama Chellappa
Lastly, we show that our end-to-end thermal-to-visible face verification system provides strong performance on the MILAB-VTF(B) dataset.
1 code implementation • 19 Jul 2021 • Vibashan VS, Jeya Maria Jose Valanarasu, Poojan Oza, Vishal M. Patel
Furthermore, we show the effectiveness of the proposed ST fusion strategy with an ablation analysis.
no code implementations • 17 Jul 2021 • Xing Di, Shuowen Hu, Vishal M. Patel
We propose a domain agnostic learning-based generative adversarial network (DAL-GAN) which can synthesize frontal views in the visible domain from thermal faces with pose variations.
1 code implementation • 14 Jul 2021 • Velat Kilic, Deepti Hegde, Vishwanath Sindagi, A. Brinton Cooper, Mark A. Foster, Vishal M. Patel
Lidar-based object detectors are critical parts of the 3D perception pipeline in autonomous navigation systems such as self-driving cars.
1 code implementation • 6 Jul 2021 • Wele Gedara Chaminda Bandara, Jeya Maria Jose Valanarasu, Vishal M. Patel
To estimate the PAN image of the up-sampled HSI, we also propose a learnable spectral response function (SRF).
Ranked #1 on
Image Super-Resolution
on Chikusei Dataset
no code implementations • 16 Jun 2021 • Pengfei Guo, Jeya Maria Jose Valanarasu, Puyang Wang, Jinyuan Zhou, Shanshan Jiang, Vishal M. Patel
Reconstructing magnetic resonance (MR) images from undersampled data is a challenging problem due to various artifacts introduced by the under-sampling operation.
no code implementations • 27 May 2021 • Poojan Oza, Vishwanath A. Sindagi, Vibashan VS, Vishal M. Patel
Recent advances in deep learning have led to the development of accurate and efficient models for various computer vision applications such as classification, segmentation, and detection.
no code implementations • 14 Apr 2021 • Rui Shao, Pramuditha Perera, Pong C. Yuen, Vishal M. Patel
A face presentation attack detection model with good generalization can be obtained when it is trained with face images from different input distributions and different types of spoof attacks.
no code implementations • 14 Apr 2021 • Poojan Oza, Vishal M. Patel
Using FL/SL frameworks, we can alleviate the lack of negative data problem by training a user authentication model over multiple user data distributed across devices.
1 code implementation • 13 Apr 2021 • Rakhil Immidisetti, Shuowen Hu, Vishal M. Patel
Existing thermal-to-visible face verification approaches expect the thermal and visible face images to be of similar resolution.
1 code implementation • 9 Apr 2021 • Xing Di, Vishal M. Patel
Extensive experiments and comparisons with several state-of-the-art methods are performed to verify the effectiveness of the proposed attribute-based multimodal synthesis method.
no code implementations • CVPR 2021 • Vibashan VS, Vikram Gupta, Poojan Oza, Vishwanath A. Sindagi, Vishal M. Patel
Existing approaches for unsupervised domain adaptive object detection perform feature alignment via adversarial training.
1 code implementation • CVPR 2021 • Pengfei Guo, Puyang Wang, Jinyuan Zhou, Shanshan Jiang, Vishal M. Patel
However, the generalizability of models trained with the FL setting can still be suboptimal due to domain shift, which results from the data collected at multiple institutions with different sensors, disease types, and acquisition protocols, etc.
2 code implementations • 21 Feb 2021 • Jeya Maria Jose Valanarasu, Poojan Oza, Ilker Hacihaliloglu, Vishal M. Patel
The proposed Medical Transformer (MedT) is evaluated on three different medical image segmentation datasets and it is shown that it achieves better performance than the convolutional and other related transformer-based architectures.
Ranked #1 on
Medical Image Segmentation
on Brain US
1 code implementation • 23 Jan 2021 • Shao-Yuan Lo, Vishal M. Patel
In this paper, we propose a new image transformation defense based on error diffusion halftoning, and combine it with adversarial training to defend against adversarial examples.
no code implementations • 8 Jan 2021 • Pramuditha Perera, Poojan Oza, Vishal M. Patel
One-Class Classification (OCC) is a special case of multi-class classification, where data observed during training is from a single positive class.
no code implementations • 7 Jan 2021 • Domenick Poster, Matthew Thielke, Robert Nguyen, Srinivasan Rajaraman, Xing Di, Cedric Nimpa Fondje, Vishal M. Patel, Nathaniel J. Short, Benjamin S. Riggan, Nasser M. Nasrabadi, Shuowen Hu
Thermal face imagery, which captures the naturally emitted heat from the face, is limited in availability compared to face imagery in the visible spectrum.
1 code implementation • ICCV 2021 • Yu Zeng, Zhe Lin, Huchuan Lu, Vishal M. Patel
The auxiliary branch (i. e. CR loss) is required only during training, and only the inpainting generator is required during the inference.
Ranked #10 on
Image Inpainting
on Places2
1 code implementation • 8 Dec 2020 • Shao-Yuan Lo, Jeya Maria Jose Valanarasu, Vishal M. Patel
Adversarial robustness of deep neural networks is an extensively studied problem in the literature and various methods have been proposed to defend against adversarial images.
1 code implementation • 25 Nov 2020 • Yu Zeng, Zhe Lin, Huchuan Lu, Vishal M. Patel
Due to the lack of supervision signals for the correspondence between missing regions and known regions, it may fail to find proper reference features, which often leads to artifacts in the results.
1 code implementation • 16 Nov 2020 • Jeya Maria Jose Valanarasu, Vishal M. Patel
This method uses undercomplete representations of the input data which makes it not so robust and more dependent on pre-training.
no code implementations • 4 Nov 2020 • He Zhang, Jianming Zhang, Federico Perazzi, Zhe Lin, Vishal M. Patel
In this paper, we propose a new method which can automatically generate high-quality image compositing without any user input.
1 code implementation • 20 Oct 2020 • Rajeev Yasarla, Jeya Maria Jose Valanarasu, Vishal M. Patel
Removal of rain streaks from a single image is an extremely challenging problem since the rainy images often contain rain streaks of different size, shape, direction and density.
1 code implementation • 4 Oct 2020 • Jeya Maria Jose Valanarasu, Vishwanath A. Sindagi, Ilker Hacihaliloglu, Vishal M. Patel
To overcome this issue, we propose using an overcomplete convolutional architecture where we project our input image into a higher dimension such that we constrain the receptive field from increasing in the deep layers of the network.
Ranked #1 on
Medical Image Segmentation
on RITE
no code implementations • 17 Sep 2020 • Shao-Yuan Lo, Vishal M. Patel
In this paper, we propose a novel attack method against video recognition models, Multiplicative Adversarial Videos (MultAV), which imposes perturbation on video data by multiplication.
1 code implementation • 14 Sep 2020 • Deepak Babu Sam, Abhinav Agarwalla, Jimmy Joseph, Vishwanath A. Sindagi, R. Venkatesh Babu, Vishal M. Patel
Dense crowd counting is a challenging task that demands millions of head annotations for training models.
no code implementations • 11 Sep 2020 • Shao-Yuan Lo, Vishal M. Patel
With a multiple BN structure, each BN brach is responsible for learning the distribution of a single perturbation type and thus provides more precise distribution estimations.
1 code implementation • ECCV 2020 • Rui Shao, Pramuditha Perera, Pong C. Yuen, Vishal M. Patel
In this paper, we show that open-set recognition systems are vulnerable to adversarial attacks.
1 code implementation • 6 Aug 2020 • Pengfei Guo, Puyang Wang, Rajeev Yasarla, Jinyuan Zhou, Vishal M. Patel, Shanshan Jiang
Data-driven automatic approaches have demonstrated their great potential in resolving various clinical diagnostic dilemmas in neuro-oncology, especially with the help of standard anatomic and advanced molecular MR images.
no code implementations • 16 Jul 2020 • Rajeev Yasarla, Vishal M. Patel
Atmospheric turbulence significantly affects imaging systems which use light that has propagated through long atmospheric paths.
2 code implementations • 11 Jul 2020 • Yashasvi Baweja, Poojan Oza, Pramuditha Perera, Vishal M. Patel
Anomaly detection-based spoof attack detection is a recent development in face Presentation Attack Detection (fPAD), where a spoof detector is learned using only non-attacked images of users.
no code implementations • ECCV 2020 • Vishwanath A. Sindagi, Rajeev Yasarla, Deepak Sam Babu, R. Venkatesh Babu, Vishal M. Patel
In this work, we focus on reducing the annotation efforts by learning to count in the crowd from limited number of labeled samples while leveraging a large pool of unlabeled data.
1 code implementation • 26 Jun 2020 • Pengfei Guo, Puyang Wang, Jinyuan Zhou, Vishal M. Patel, Shanshan Jiang
Data-driven automatic approaches have demonstrated their great potential in resolving various clinical diagnostic dilemmas for patients with malignant gliomas in neuro-oncology with the help of conventional and advanced molecular MR images.
no code implementations • 21 Jun 2020 • Pramuditha Perera, Julian Fierrez, Vishal M. Patel
In this paper, we investigate how to detect intruders with low latency for Active Authentication (AA) systems with multiple-users.
3 code implementations • 8 Jun 2020 • Jeya Maria Jose, Vishwanath Sindagi, Ilker Hacihaliloglu, Vishal M. Patel
Due to its excellent performance, U-Net is the most widely used backbone architecture for biomedical image segmentation in the recent years.
no code implementations • 29 May 2020 • Rui Shao, Pramuditha Perera, Pong C. Yuen, Vishal M. Patel
A face presentation attack detection model with good generalization can be obtained when it is trained with face images from different input distributions and different types of spoof attacks.
no code implementations • 20 Apr 2020 • Xing Di, Benjamin S. Riggan, Shuowen Hu, Nathaniel J. Short, Vishal M. Patel
Finally, a pre-trained VGG-Face network is leveraged to extract features from the synthesized image and the input visible image for verification.
no code implementations • 7 Apr 2020 • Vishwanath A. Sindagi, Rajeev Yasarla, Vishal M. Patel
The proposed Confidence Guided Deep Residual Counting Network (CG-DRCN) is evaluated on recent complex datasets, and it achieves significant improvements in errors.
no code implementations • 18 Dec 2019 • Jeya Maria Jose V., Rajeev Yasarla, Puyang Wang, Ilker Hacihaliloglu, Vishal M. Patel
We show that our method can synthesize high-quality US images for every manipulated segmentation label with qualitative and quantitative improvements over the recent state-of-the-art synthesis methods.
no code implementations • 17 Dec 2019 • Xing Di, Vishal M. Patel
In this paper, we take a different approach, where we formulate the original problem as a stage-wise learning problem.
no code implementations • ECCV 2020 • Vishwanath A. Sindagi, Poojan Oza, Rajeev Yasarla, Vishal M. Patel
Adverse weather conditions such as haze and rain corrupt the quality of captured images, which cause detection networks trained on clean images to perform poorly on these images.
no code implementations • ICCV 2019 • Vishwanath A. Sindagi, Rajeev Yasarla, Vishal M. Patel
The proposed Confidence Guided Deep Residual Counting Network (CG-DRCN) is evaluated on recent complex datasets, and it achieves significant improvements in errors.
no code implementations • 10 Sep 2019 • Rajeev Yasarla, Vishal M. Patel
Single image de-raining is an extremely challenging problem since the rainy images contain rain streaks which often vary in size, direction and density.