Search Results for author:
Found 37 papers, 9 papers with code
Zero-Shot Knowledge Distillation in Deep Networks
Without even using any meta-data, we synthesize the Data Impressions from the complex Teacher model and utilize these as surrogates for the original training data samples to transfer its learning to Student via knowledge distillation.
Sketch-Guided Object Localization in Natural Images
We refer to this problem as sketch-guided object localization.
A Context-aware Delayed Agglomeration Framework for Electron Microscopy Segmentation
In particular, given an over-segmented image or volume, we propose a novel framework for accurately clustering regions of the same neuron.
CoNMix for Source-free Single and Multi-target Domain Adaptation
The main motive of this work is to solve for Single and Multi target Domain Adaptation (SMTDA) for the source-free paradigm, which enforces a constraint where the labeled source data is not available during target adaptation due to various privacy-related restrictions on data sharing.
MMD-ReID: A Simple but Effective Solution for Visible-Thermal Person ReID
Learning modality invariant features is central to the problem of Visible-Thermal cross-modal Person Reidentification (VT-ReID), where query and gallery images come from different modalities.
Understanding the Galactic population of normal pulsars: A leap forward
By comparing the distributions of various parameters of synthetic pulsars detectable by the Parkes Multibeam Pulsar Survey, the Pulsar Arecibo L-band Feed Array Survey, and two Swinburne Multibeam surveys with those of the real pulsars detected by the same surveys, we find that a good and physically realistic model can be obtained by using a uniform distribution of the braking index in the range of 2. 5 to 3. 0, a uniform distribution of the cosine of the angle between the spin and the magnetic axis in the range of 0 to 1, a log-normal birth distribution of the surface magnetic field with the mean and the standard deviation as 12. 85 and 0. 55 respectively while keeping the distributions of other parameters unchanged from the ones most commonly used in the literature.
High Energy Astrophysical Phenomena Solar and Stellar Astrophysics
Robust Few-shot Learning Without Using any Adversarial Samples
These methods rely on the generation of adversarial samples in every episode of training, which further adds a computational burden.
Data-free Defense of Black Box Models Against Adversarial Attacks
At test time, WNR combined with trained regenerator network is prepended to the black box network, resulting in a high boost in adversarial accuracy.
DAD++: Improved Data-free Test Time Adversarial Defense
With the increasing deployment of deep neural networks in safety-critical applications such as self-driving cars, medical imaging, anomaly detection, etc., adversarial robustness has become a crucial concern in the reliability of these networks in real-world scenarios.
Operator-in-the-Loop Deep Sequential Multi-camera Feature Fusion for Person Re-identification
Therefore, target identifications by operator in a subset of cameras cannot be utilized to improve ranking of the target in remaining set of network cameras.
Adversarial Attacks and Defences: A Survey
Deep learning has emerged as a strong and efficient framework that can be applied to a broad spectrum of complex learning problems which were difficult to solve using the traditional machine learning techniques in the past.
Enhancing Salient Object Segmentation Through Attention
Segmenting salient objects in an image is an important vision task with ubiquitous applications.
EvAn: Neuromorphic Event-based Anomaly Detection
Event-based cameras are bio-inspired novel sensors that asynchronously record changes in illumination in the form of events, thus resulting in significant advantages over conventional cameras in terms of low power utilization, high dynamic range, and no motion blur.
DeGAN : Data-Enriching GAN for Retrieving Representative Samples from a Trained Classifier
We use the available data, that may be an imbalanced subset of the original training dataset, or a related domain dataset, to retrieve representative samples from a trained classifier, using a novel Data-enriching GAN (DeGAN) framework.
Semantic Segmentation of highly class imbalanced fully labelled 3D volumetric biomedical images and unsupervised Domain Adaptation of the pre-trained Segmentation Network to segment another fully unlabelled Biomedical 3D Image stack
We first perform Semantic Segmentation on the fully labeled isotropic biomedical source data (FIBSEM) and try to incorporate the the trained model for segmenting the target unlabelled dataset(SNEMI3D)which shares some similarities with the source dataset in the context of different types of cellular bodies and other cellular components.
Self-Supervised 3D Human Pose Estimation via Part Guided Novel Image Synthesis
Camera captured human pose is an outcome of several sources of variation.
Kinematic-Structure-Preserved Representation for Unsupervised 3D Human Pose Estimation
However, generalizability of human pose estimation models developed using supervision on large-scale in-studio datasets remains questionable, as these models often perform unsatisfactorily on unseen in-the-wild environments.
Kernel Density Estimation based Factored Relevance Model for Multi-Contextual Point-of-Interest Recommendation
An automated contextual suggestion algorithm is likely to recommend contextually appropriate and personalized 'points-of-interest' (POIs) to a user, if it can extract information from the user's preference history (exploitation) and effectively blend it with the user's current contextual information (exploration) to predict a POI's 'appropriateness' in the current context.
Fusion of Deep and Non-Deep Methods for Fast Super-Resolution of Satellite Images
In the emerging commercial space industry there is a drastic increase in access to low cost satellite imagery.
Text-based Person Search via Attribute-aided Matching
Text-based person search aims to retrieve the pedestrian images that best match a given text query.
Ranked #17 on
Text based Person Retrieval
on CUHK-PEDES
Effectiveness of Arbitrary Transfer Sets for Data-free Knowledge Distillation
In such scenarios, existing approaches either iteratively compose a synthetic set representative of the original training dataset, one sample at a time or learn a generative model to compose such a transfer set.
Single Image Super-resolution with a Switch Guided Hybrid Network for Satellite Images
We will dive into the recent evolution of the deep models in the context of SISR over the past few years and will present a comparative study between these models.
Mining Data Impressions from Deep Models as Substitute for the Unavailable Training Data
We dub them "Data Impressions", which act as proxy to the training data and can be used to realize a variety of tasks.
Incremental Learning for Animal Pose Estimation using RBF k-DPP
Thus, in this work, we propose a novel problem of "Incremental Learning for Animal Pose Estimation".
Beyond Classification: Knowledge Distillation using Multi-Object Impressions
Knowledge Distillation (KD) utilizes training data as a transfer set to transfer knowledge from a complex network (Teacher) to a smaller network (Student).
Uncertainty-Aware Adaptation for Self-Supervised 3D Human Pose Estimation
The advances in monocular 3D human pose estimation are dominated by supervised techniques that require large-scale 2D/3D pose annotations.
Ranked #8 on
Unsupervised 3D Human Pose Estimation
on Human3.6M
Monocular 3D Human Pose Estimation
Unsupervised 3D Human Pose Estimation
+2
DAD: Data-free Adversarial Defense at Test Time
Deep models are highly susceptible to adversarial attacks.
Non-Local Latent Relation Distillation for Self-Adaptive 3D Human Pose Estimation
To this end, we cast 3D pose learning as a self-supervised adaptation problem that aims to transfer the task knowledge from a labeled source domain to a completely unpaired target.
Ranked #5 on
Unsupervised 3D Human Pose Estimation
on Human3.6M
Holistic Approach to Measure Sample-level Adversarial Vulnerability and its Utility in Building Trustworthy Systems
We, therefore, propose a holistic approach for quantifying adversarial vulnerability of a sample by combining these different perspectives, i. e., degree of model's reliance on high-frequency features and the (conventional) sample-distance to the decision boundary.
DE-CROP: Data-efficient Certified Robustness for Pretrained Classifiers
Existing works use this technique to provably secure a pretrained non-robust model by training a custom denoiser network on entire training data.
Grounding Scene Graphs on Natural Images via Visio-Lingual Message Passing
In VL-MPAG Net, we first construct a directed graph with object proposals as nodes and an edge between a pair of nodes representing a plausible relation between them.
Robustifying Deep Vision Models Through Shape Sensitization
We also obtain gains of up to 28% and 8. 5% on natural adversarial and out-of-distribution datasets like ImageNet-A (for ViT-B) and ImageNet-R (for ViT-S), respectively.
Multimodal Query-guided Object Localization
In such a scenario, a hand-drawn sketch of the object could be a choice for a query.
Query-guided Attention in Vision Transformers for Localizing Objects Using a Single Sketch
and Sketchy datasets, respectively, and a $12. 2\%$ improvement in AP@50 for large objects that are `unseen' during training.
Edges to Shapes to Concepts: Adversarial Augmentation for Robust Vision
We show that our augmentations significantly improve classification accuracy and robustness measures on a range of datasets and neural architectures.
Adaptive Self-Distillation for Minimizing Client Drift in Heterogeneous Federated Learning
In practice, there can often be substantial heterogeneity (e. g., class imbalance) across the local data distributions observed by each of these clients.
Scalable Model-Based Gaussian Process Clustering
Gaussian process is an indispensable tool in clustering functional data, owing to it's flexibility and inherent uncertainty quantification.
