A5013112157- Advanced Vision and Imaging
- Image Enhancement Techniques
- Advanced Image Processing Techniques
- Video Surveillance and Tracking Methods
- Natural Language Processing Techniques
- Spam and Phishing Detection
- Remote Sensing and LiDAR Applications
- Misinformation and Its Impacts
- Image and Video Quality Assessment
- Robotics and Sensor-Based Localization
- Computer Graphics and Visualization Techniques
- Visual Attention and Saliency Detection
- Authorship Attribution and Profiling
- Topic Modeling
- COVID-19 diagnosis using AI
- Animal Vocal Communication and Behavior
- Speech Recognition and Synthesis
- Image Processing Techniques and Applications
- Advanced Image and Video Retrieval Techniques
- Advanced Neural Network Applications
- Wildlife Ecology and Conservation
- Robotic Path Planning Algorithms
- Adversarial Robustness in Machine Learning
- Color Science and Applications
- Identification and Quantification in Food
Jadavpur University
2018-2022
Shanghai Artificial Intelligence Laboratory
2022
Indian Institute of Technology Madras
2020
Huawei Technologies (Sweden)
2019
This paper reviews the NTIRE 2020 Challenge on Non-Homogeneous Dehazing of images (restoration rich details in hazy image). We focus proposed solutions and their results evaluated NH-Haze, a novel dataset consisting 55 pairs real haze free nonhomogeneous recorded outdoor. NH-Haze is first realistic that provides ground truth images. The has been produced using professional generator imitates conditions scenes. 168 participants registered challenge 27 teams competed final testing phase. gauge...
This paper reports on the NTIRE 2022 challenge perceptual image quality assessment (IQA), held in conjunction with New Trends Image Restoration and Enhancement workshop (NTIRE) at CVPR 2022. is to address emerging of IQA by processing algorithms. The output images these algorithms have completely different characteristics from traditional distortions are included PIPAL dataset used this challenge. divided into two tracks, a full-reference track similar previous new that focuses no-reference...
Recently, CNN based end-to-end deep learning methods achieve superiority in Image Dehazing but they tend to fail drastically Non-homogeneous dehazing. Apart from that, existing popular Multi-scale approaches are runtime intensive and memory inefficient. In this context, we proposed a fast Deep Multi-patch Hierarchical Network restore hazed images by aggregating features multiple image patches different spatial sections of the with fewer number network parameters. Our method is quite robust...
Motion blur is one of the most common degradation artifacts in dynamic scene photography. This paper reviews NTIRE 2020 Challenge on Image and Video Deblurring. In this challenge, we present evaluation results from 3 competition tracks as well proposed solutions. Track 1 aims to develop single-image deblurring methods focusing restoration quality. On 2, image are executed a mobile platform find balance running speed accuracy. targets developing video that exploit temporal relation between...
Super-Resolution (SR) is a fundamental computer vision task that aims to obtain high-resolution clean image from the given low-resolution counterpart. This paper reviews NTIRE 2021 Challenge on Video Super-Resolution. We present evaluation results two competition tracks as well proposed solutions. Track 1 develop conventional video SR methods focusing restoration quality. 2 assumes more challenging environment with lower frame rates, casting spatio-temporal problem. In each competition, 247...
Camera scene detection is among the most popular computer vision problem on smartphones. While many custom solutions were developed for this task by phone vendors, none of designed models available publicly up until now. To address problem, we introduce first Mobile AI challenge, where target to develop quantized deep learning-based camera classification that can demonstrate a real-time performance smartphones and IoT platforms. For this, participants provided with large-scale CamSDD dataset...
The Bokeh Effect is one of the most desirable effects in photography for rendering artistic and aesthetic photos. Usually, it requires a DSLR camera with different aperture shutter settings certain skills to generate this effect. In smartphones, computational methods additional sensors are used overcome physical lens sensor limitations achieve such Most existing utilized sensor's data or pretrained network fine depth estimation scene sometimes use portrait segmentation module segment salient...
This paper reviews the first-ever image demoireing challenge that was part of Advances in Image Manipulation (AIM) workshop, held conjunction with ICCV 2019. describes challenge, and focuses on proposed solutions their results. Demoireing is a difficult task removing moire patterns from an to reveal underlying clean image. A new dataset, called LCDMoire created for this consists 10,200 synthetically generated pairs (moire ground truth). The divided into 2 tracks. Track 1 targeted fidelity,...
Custom and natural lighting conditions can be emulated in images of the scene during post-editing. Extraordinary capabilities deep learning framework utilized for such purpose. Deep image relighting allows automatic photo enhancement by illumination-specific retouching. Most state-of-the-art methods are run-time intensive memory inefficient. In this paper, we propose an efficient, real-time Stacked Relighting Network (DSRN) utilizing aggregated features from input at different scales. Our...
The area of Handwritten Signature Verification has been broadly researched in the last decades, but remains an open research problem. In offline (static) signature verification, dynamic information writing process is lost, and it difficult to design good feature extractors that can distinguish genuine signatures skilled forgeries. This verification task even harder writer independent scenarios which undeniably fiscal for realistic cases. this paper, we have proposed Ensemble model writer,...
This paper reviews the second AIM realistic bokeh effect rendering challenge and provides description of proposed solutions results. The participating teams were solving a real-world simulation problem, where goal was to learn shallow focus technique using large-scale EBB! dataset consisting 5K / wide depth-of-field image pairs captured Canon 7D DSLR camera. participants had render based on only one single frame without any additional data from other cameras or sensors. target metric used in...
Deep image relighting allows photo enhancement by illumination-specific retouching without human effort and so it is getting much interest lately. Most of the existing popular methods available for are run-time intensive memory inefficient. Keeping these issues in mind, we propose use Stacked Multi-Scale Hierarchical Network, which aggregates features from each at different scales. Our solution differentiable robust translating illumination setting input to target image. Additionally, have...
Language identification of social media text still remains a challenging task due to properties like code-mixing and inconsistent phonetic transliterations. In this paper, we present supervised learning approach for language at the word level low resource Bengali-English code-mixed data taken from media. We employ two methods encoding, namely character based root phone train our deep LSTM models. Utilizing these models created ensemble using stacking threshold technique which gave 91.78%...
We review the AIM 2020 challenge on virtual image relighting and illumination estimation. This paper presents novel VIDIT dataset used in different proposed solutions final evaluation results over 3 tracks. The first track considered one-to-one relighting; objective was to relight an input photo of a scene with color temperature illuminant orientation (i.e., light source position). goal second estimate settings, namely orientation, from given image. Lastly, third dealt any-to-any relighting,...
Convolutional Neural Network models have successfully detected retinal illness from optical coherence tomography (OCT) and fundus images. These CNN frequently rely on vast amounts of labeled data for training, difficult to obtain, especially rare diseases. Furthermore, a deep learning system trained set with only one or few diseases cannot detect other diseases, limiting the system's practical use in disease identification. We introduced an unsupervised approach detecting anomalies images...
Development of perceptual image quality assessment (IQA) metrics has been significant interest to computer vision community. The aim these is model an as perceived by humans. Recent works in Full-reference IQA research perform pixelwise comparison between deep features corresponding query and reference images for prediction. However, feature may not be meaningful if distortion present severe. In this context, we explore utilization no-reference task. Our consists both full-reference...
The objective of this survey paper is finding methodology for estimation air pollution that comes from road traffic. A detailed study has been done pollution. In paper, we have studied relevant papers and designed a near to the movement on-road vehicles. This research important verification national ambient quality standards (NAAQS), especially at living places. Because NAAQ should follow any place where people live in consideration health effects related traffic-related