Search Results for author:
Found 51 papers, 21 papers with code
An Inference Algorithm for Multi-Label MRF-MAP Problems with Clique Size 100
Earlier algorithms based on this transformation could not handle problems larger than 16 labels on cliques of size 4.
Is Sharing of Egocentric Video Giving Away Your Biometric Signature?
In this work, we create a novel kind of privacy attack by extracting the wearer’s gait profile, a well known biometric signature, from such optical flow in the egocentric videos.
Model Generation from Requirements with LLMs: an Exploratory Study
Complementing natural language (NL) requirements with graphical models can improve stakeholders' communication and provide directions for system design.
ECoDepth: Effective Conditioning of Diffusion Models for Monocular Depth Estimation
We argue that the embedding vector from a ViT model, pre-trained on a large dataset, captures greater relevant information for SIDE than the usual route of generating pseudo image captions, followed by CLIP based text embeddings.
Ranked #3 on
Monocular Depth Estimation
on NYU-Depth V2
FocusMAE: Gallbladder Cancer Detection from Ultrasound Videos with Focused Masked Autoencoders
We validate the proposed methods on the curated dataset, and report a new state-of-the-art (SOTA) accuracy of 96. 4% for the GBC detection problem, against an accuracy of 84% by current Image-based SOTA - GBCNet, and RadFormer, and 94. 7% by Video-based SOTA - AdaMAE.
A Study of Fairness Concerns in AI-based Mobile App Reviews
Finally, the manual analysis of 2, 248 app owners' responses to the fairness reviews identified six root causes (e. g., 'copyright issues') that app owners report to justify fairness concerns.
Army of Thieves: Enhancing Black-Box Model Extraction via Ensemble based sample selection
In this work, we explore the usage of an ensemble of deep learning models as our thief model.
United We Stand, Divided We Fall: UnityGraph for Unsupervised Procedure Learning from Videos
Given multiple videos of the same task, procedure learning addresses identifying the key-steps and determining their order to perform the task.
Model-driven Engineering for Machine Learning Components: A Systematic Literature Review
Through this SLR, we wanted to analyze existing studies, including their motivations, MDE solutions, evaluation techniques, key benefits and limitations.
Gall Bladder Cancer Detection from US Images with Only Image Level Labels
We posit that even when we have only the image level label, still formulating the problem as object detection (with bounding box output) helps a deep neural network (DNN) model focus on the relevant region of interest.
UTRNet: High-Resolution Urdu Text Recognition In Printed Documents
To address the limitations of previous works, which struggle to generalize to the intricacies of the Urdu script and the lack of sufficient annotated real-world data, we have introduced the UTRSet-Real, a large-scale annotated real-world dataset comprising over 11, 000 lines and UTRSet-Synth, a synthetic dataset with 20, 000 lines closely resembling real-world and made corrections to the ground truth of the existing IIITH dataset, making it a more reliable resource for future research.
Ranked #1 on
Printed Text Recognition
on UPTI
Stop Words for Processing Software Engineering Documents: Do they Matter?
In this work, we investigate the usefulness of stop word removal in a software engineering context.
Requirements Engineering Framework for Human-centered Artificial Intelligence Software Systems
[Method] In this paper, we present a new framework developed based on human-centered AI guidelines and a user survey to aid in collecting requirements for human-centered AI-based software.
From Forks to Forceps: A New Framework for Instance Segmentation of Surgical Instruments
Minimally invasive surgeries and related applications demand surgical tool classification and segmentation at the instance level.
RadFormer: Transformers with Global-Local Attention for Interpretable and Accurate Gallbladder Cancer Detection
We propose a novel deep neural network architecture to learn interpretable representation for medical image analysis.
Attention Attention Everywhere: Monocular Depth Prediction with Skip Attention
Typically, a skip connection module is used to fuse the encoder and decoder features, which comprises of feature map concatenation followed by a convolution operation.
Ranked #18 on
Monocular Depth Estimation
on KITTI Eigen split
(using extra training data)
Reducing Annotation Effort by Identifying and Labeling Contextually Diverse Classes for Semantic Segmentation Under Domain Shift
In this work, we propose an ADA strategy, which given a frame, identifies a set of classes that are hardest for the model to predict accurately, thereby recommending semantically meaningful regions to be annotated in a selected frame.
A Novel Data Augmentation Technique for Out-of-Distribution Sample Detection using Compounded Corruptions
One of the major advantages of CnC is that it does not require any hold-out data apart from the training set.
Unsupervised Contrastive Learning of Image Representations from Ultrasound Videos with Hard Negative Mining
We further validate the generalizability of our method on a publicly available lung US image dataset of COVID-19 pathologies and show an improvement of 1. 5% compared to SOTA.
My View is the Best View: Procedure Learning from Egocentric Videos
Instead, we propose to use the signal provided by the temporal correspondences between key-steps across videos.
Depthformer : Multiscale Vision Transformer For Monocular Depth Estimation With Local Global Information Fusion
We also propose a Transbins module that divides the depth range into bins whose center value is estimated adaptively per image.
Ranked #28 on
Monocular Depth Estimation
on KITTI Eigen split
(using extra training data)
TAPHSIR: Towards AnaPHoric Ambiguity Detection and ReSolution In Requirements
We introduce TAPHSIR, a tool for anaphoric ambiguity detection and anaphora resolution in requirements.
Surpassing the Human Accuracy: Detecting Gallbladder Cancer from USG Images with Curriculum Learning
However, USG images are challenging to analyze due to low image quality, noise, and varying viewpoints due to the handheld nature of the sensor.
Ranked #1 on
Gallbladder Cancer Detection
on GBCU
A Stitch in Time Saves Nine: A Train-Time Regularizing Loss for Improved Neural Network Calibration
We show that training with MDCA leads to better-calibrated models in terms of Expected Calibration Error ( ECE ), and Static Calibration Error ( SCE ) on image classification, and segmentation tasks.
Adversarial Attacks on Speech Recognition Systems for Mission-Critical Applications: A Survey
Indeed, Adversarial Artificial Intelligence (AI) which refers to a set of techniques that attempt to fool machine learning models with deceptive data, is a growing threat in the AI and machine learning research community, in particular for machine-critical applications.
To miss-attend is to misalign! Residual Self-Attentive Feature Alignment for Adapting Object Detectors
Qualitative results demonstrate the ability of ILLUME to attend important object instances required for alignment.
Merry Go Round: Rotate a Frame and Fool a DNN
On the other hand DNNs are known to be susceptible to Adversarial Attacks (AAs) which add im-perceptible noise to the input.
Attention Guided Cosine Margin For Overcoming Class-Imbalance in Few-Shot Road Object Detection
To overcome these pitfalls in metric learning based FSOD techniques, we introduce Attention Guided Cosine Margin (AGCM) that facilitates the creation of tighter and well separated class-specific feature clusters in the classification head of the object detector.
Multi-Domain Incremental Learning for Semantic Segmentation
Recent efforts in multi-domain learning for semantic segmentation attempt to learn multiple geographical datasets in a universal, joint model.
Does Data Repair Lead to Fair Models? Curating Contextually Fair Data To Reduce Model Bias
Through a series of experiments, we validate that curating contextually fair data helps make model predictions fair by balancing the true positive rate for the protected class across groups without compromising on the model's overall performance.
Efficient and Generic Interactive Segmentation Framework to Correct Mispredictions during Clinical Evaluation of Medical Images
We report a time saving of 2. 8, 3. 0, 1. 9, 4. 4, and 8. 6 fold compared to other interactive segmentation techniques.
Anonymizing Egocentric Videos
In a more damaging scenario, one can even recognize a wearer using hand gestures from egocentric videos, or identify a wearer in third person videos such as from a surveillance camera.
Contextual Diversity for Active Learning
Contextual Diversity (CD) hinges on a crucial observation that the probability vector predicted by a CNN for a region of interest typically contains information from a larger receptive field.
CovidAID: COVID-19 Detection Using Chest X-Ray
This may be useful in an inpatient setting where the present systems are struggling to decide whether to keep the patient in the ward along with other patients or isolate them in COVID-19 areas.
Diversity in Fashion Recommendation using Semantic Parsing
In this paper, we propose to recommend images by explicitly learning and exploiting part based similarity.
Making Third Person Techniques Recognize First-Person Actions in Egocentric Videos
Unlike third person domain, researchers have divided first-person actions into two categories: involving hand-object interactions and the ones without, and developed separate techniques for the two action categories.
A Novel Technique for Evidence based Conditional Inference in Deep Neural Networks via Latent Feature Perturbation
Designing such a network, as well as collecting jointly labeled data for training is a non-trivial task.
U-SegNet: Fully Convolutional Neural Network based Automated Brain tissue segmentation Tool
Automated brain tissue segmentation into white matter (WM), gray matter (GM), and cerebro-spinal fluid (CSF) from magnetic resonance images (MRI) is helpful in the diagnosis of neuro-disorders such as epilepsy, Alzheimer's, multiple sclerosis, etc.
Inference in Higher Order MRF-MAP Problems With Small and Large Cliques
Two popular combinatorial approaches for solving such formulations are flow based and polyhedral approaches.
A Joint 3D-2D based Method for Free Space Detection on Roads
Finally, we use the obtained road segmentation with the 3D depth data from monocular SLAM to detect the free space for the navigation purposes.
Robust Monocular SLAM for Egocentric Videos
The incremental nature of SOTA SLAM, in the presence of unreliable pose and 3D estimates in egocentric videos, with no opportunities for global loop closures, generates drifts and leads to the eventual failures of such techniques.
Computing Egomotion with Local Loop Closures for Egocentric Videos
Finding the camera pose is an important step in many egocentric video applications.
Min Norm Point Algorithm for Higher Order MRF-MAP Inference
Current state of the art inference algorithms for general submodular function takes many hours for problems with clique size 16, and fail to scale beyond.
First Person Action Recognition Using Deep Learned Descriptors
It can also be trained from relatively small number of labeled egocentric videos that are available.
EgoSampling: Wide View Hyperlapse from Egocentric Videos
However, this accentuates the shake caused by natural head motion in an egocentric video, making the fast forwarded video useless.
Trajectory Aligned Features For First Person Action Recognition
Objects present in the scene and hand gestures of the wearer are the most important cues for first person action recognition but are difficult to segment and recognize in an egocentric video.
Compact CNN for Indexing Egocentric Videos
Furthermore, our CNN is able to recognize whether a video is egocentric or not with 99. 2% accuracy, up by 24% from current state-of-the-art.
EgoSampling: Fast-Forward and Stereo for Egocentric Videos
While egocentric cameras like GoPro are gaining popularity, the videos they capture are long, boring, and difficult to watch from start to end.
Temporal Segmentation of Egocentric Videos
Two sources of information for video segmentation are (i) the motion of the camera wearer, and (ii) the objects and activities recorded in the video.
Multi Label Generic Cuts: Optimal Inference in Multi Label Multi Clique MRF-MAP Problems
We exploit sparseness in the feasible configurations of the transformed 2-label problem to suggest an improvement to Generic Cuts [3] to solve the 2-label problems efficiently.
Fast Approximate Inference in Higher Order MRF-MAP Labeling Problems
Generic Cuts (GC) of Arora et al. [9] shows that when potentials are submodular, inference problems can be solved optimally in polynomial time for fixed size cliques.
