Search Results for author:
Found 109 papers, 29 papers with code
Sparse Points to Dense Clouds: Enhancing 3D Detection with Limited LiDAR Data
Our method requires only a small number of 3D points, that can be obtained from a low-cost, low-resolution sensor.
Severity Controlled Text-to-Image Generative Model Bias Manipulation
Charting the susceptibility of T2I models to such manipulation, we first expose the new possibility of a dynamic and computationally efficient exploitation of model bias by targeting the embedded language models.
Towards Temporally Consistent Referring Video Object Segmentation
Referring Video Object Segmentation (R-VOS) methods face challenges in maintaining consistent object segmentation due to temporal context variability and the presence of other visually similar objects.
Ranked #3 on
Referring Video Object Segmentation
on Refer-YouTube-VOS
(using extra training data)
3D Object Detection from Point Cloud via Voting Step Diffusion
In this work, we focus on the distributional properties of point clouds and formulate the voting process as generating new points in the high-density region of the distribution of object centers.
External Knowledge Enhanced 3D Scene Generation from Sketch
Generating realistic 3D scenes is challenging due to the complexity of room layouts and object geometries. We propose a sketch based knowledge enhanced diffusion architecture (SEK) for generating customized, diverse, and plausible 3D scenes.
Soft Masked Transformer for Point Cloud Processing with Skip Attention-Based Upsampling
This strategy allows various transformer blocks to share the same position information over the same resolution points, thereby reducing network parameters and training time without compromising accuracy. Experimental comparisons with existing methods on multiple datasets demonstrate the efficacy of SMTransformer and skip-attention-based up-sampling for point cloud processing tasks, including semantic segmentation and classification.
Beyond Skeletons: Integrative Latent Mapping for Coherent 4D Sequence Generation
To address this, we propose a novel framework that generates coherent 4D sequences with animation of 3D shapes under given conditions with dynamic evolution of shape and color over time through integrative latent mapping.
SCTransNet: Spatial-channel Cross Transformer Network for Infrared Small Target Detection
In the proposed SCTBs, the outputs of all encoders are interacted with cross transformer to generate mixed features, which are redistributed to all decoders to effectively reinforce semantic differences between the target and clutter at full scales.
Quantifying Bias in Text-to-Image Generative Models
Bias in text-to-image (T2I) models can propagate unfair social representations and may be used to aggressively market ideas or push controversial agendas.
Text-image guided Diffusion Model for generating Deepfake celebrity interactions
To curb their exploitation for Deepfakes, it is imperative to first explore the extent to which diffusion models can be used to generate realistic content that is controllable with convenient prompts.
On quantifying and improving realism of images generated with diffusion
Recent advances in diffusion models have led to a quantum leap in the quality of generative visual content.
PRAT: PRofiling Adversarial aTtacks
Inspired by this, we introduce a novel problem of PRofiling Adversarial aTtacks (PRAT).
Dual Student Networks for Data-Free Model Stealing
To this end, the two main challenges are estimating gradients of the target model without access to its parameters, and generating a diverse set of training samples that thoroughly explores the input space.
Quantum-Inspired Machine Learning: a Survey
Quantum-inspired Machine Learning (QiML) is a burgeoning field, receiving global attention from researchers for its potential to leverage principles of quantum mechanics within classical computational frameworks.
Sketch and Text Guided Diffusion Model for Colored Point Cloud Generation
In this paper, we propose a sketch and text guided probabilistic diffusion model for colored point cloud generation that conditions the denoising process jointly with a hand drawn sketch of the object and its textual description.
BAGM: A Backdoor Attack for Manipulating Text-to-Image Generative Models
Based on the penetration level, BAGM takes the form of a suite of attacks that are referred to as surface, shallow and deep attacks in this article.
Spectrum-guided Multi-granularity Referring Video Object Segmentation
To address the drift problem, we propose a Spectrum-guided Multi-granularity (SgMg) approach, which performs direct segmentation on the encoded features and employs visual details to further optimize the masks.
Ranked #1 on
Referring Expression Segmentation
on J-HMDB
(using extra training data)
A Comprehensive Overview of Large Language Models
Large Language Models (LLMs) have recently demonstrated remarkable capabilities in natural language processing tasks and beyond.
UnLoc: A Universal Localization Method for Autonomous Vehicles using LiDAR, Radar and/or Camera Input
The results ascertain the efficacy of our technique.
Human Gesture and Gait Analysis for Autism Detection
Autism diagnosis presents a major challenge due to the vast heterogeneity of the condition and the elusive nature of early detection.
DANet: Density Adaptive Convolutional Network with Interactive Attention for 3D Point Clouds
Local features and contextual dependencies are crucial for 3D point cloud analysis.
Full Point Encoding for Local Feature Aggregation in 3D Point Clouds
In particular, we achieve state-of-the-art semantic segmentation results of 76% mIoU on S3DIS 6-fold and 72. 2% on S3DIS Area5.
Q-Cogni: An Integrated Causal Reinforcement Learning Framework
We present Q-Cogni, an algorithmically integrated causal reinforcement learning framework that redesigns Q-Learning with an autonomous causal structure discovery method to improve the learning process with causal inference.
Slice Transformer and Self-supervised Learning for 6DoF Localization in 3D Point Cloud Maps
We present a self-supervised learning method that employs Transformers for the first time for the task of outdoor localization using LiDAR data.
Learning Sparse Temporal Video Mapping for Action Quality Assessment in Floor Gymnastics
Athlete performance measurement in sports videos requires modeling long sequences since the entire spatio-temporal progression contributes dominantly to the performance.
3D Spatial Multimodal Knowledge Accumulation for Scene Graph Prediction in Point Cloud
In-depth understanding of a 3D scene not only involves locating/recognizing individual objects, but also requires to infer the relationships and interactions among them.
Fast Parallel Exact Inference on Bayesian Networks: Poster
Bayesian networks (BNs) are attractive, because they are graphical and interpretable machine learning models.
Fast Parallel Bayesian Network Structure Learning
The mainstream BN structure learning methods require performing a large number of conditional independence (CI) tests.
Query Efficient Cross-Dataset Transferable Black-Box Attack on Action Recognition
By using a nearly disjoint dataset to train the substitute model, our method removes the requirement that the substitute model be trained using the same dataset as the target model, and leverages queries to the target model to retain the fooling rate benefits provided by query-based methods.
3DMODT: Attention-Guided Affinities for Joint Detection & Tracking in 3D Point Clouds
We propose a method for joint detection and tracking of multiple objects in 3D point clouds, a task conventionally treated as a two-step process comprising object detection followed by data association.
Vision Transformer Based Model for Describing a Set of Images as a Story
Visual Story-Telling is the process of forming a multi-sentence story from a set of images.
Ranked #25 on
Visual Storytelling
on VIST
Vision Transformers for Action Recognition: A Survey
Finally, it provides a discussion on the challenges, outlook, and future avenues for this research direction.
Domain-invariant Prototypes for Semantic Segmentation
In this paper, we present an easy-to-train framework that learns domain-invariant prototypes for domain adaptive semantic segmentation.
Contrastive Self-Supervised Learning Leads to Higher Adversarial Susceptibility
We establish that this is a result of the presence of false negative pairs in the training process, which increases model sensitivity to input perturbations.
Region Aware Video Object Segmentation with Deep Motion Modeling
Current semi-supervised video object segmentation (VOS) methods usually leverage the entire features of one frame to predict object masks and update memory.
Ranked #16 on
Semi-Supervised Video Object Segmentation
on DAVIS 2017 (val)
(using extra training data)
Learning from Pixel-Level Noisy Label : A New Perspective for Light Field Saliency Detection
Saliency detection with light field images is becoming attractive given the abundant cues available, however, this comes at the expense of large-scale pixel level annotated data which is expensive to generate.
Visual Attention Methods in Deep Learning: An In-Depth Survey
Inspired by the human cognitive system, attention is a mechanism that imitates the human cognitive awareness about specific information, amplifying critical details to focus more on the essential aspects of data.
UNICON: Combating Label Noise Through Uniform Selection and Contrastive Learning
To combat label noise, recent state-of-the-art methods employ some sort of sample selection mechanism to select a possibly clean subset of data.
DualConv: Dual Convolutional Kernels for Lightweight Deep Neural Networks
We also test DualConv for image detection on YOLO-V3.
Bi-CLKT: Bi-Graph Contrastive Learning based Knowledge Tracing
The goal of Knowledge Tracing (KT) is to estimate how well students have mastered a concept based on their historical learning of related exercises.
Learning From Pixel-Level Noisy Label: A New Perspective for Light Field Saliency Detection
Saliency detection with light field images is becoming attractive given the abundant cues available, however, this comes at the expense of large-scale pixel level annotated data which is expensive to generate.
Mesh Convolution with Continuous Filters for 3D Surface Parsing
In this paper, we propose a series of modular operations for effective geometric feature learning from 3D triangle meshes.
Object-to-Scene: Learning to Transfer Object Knowledge to Indoor Scene Recognition
The final results in this work show that OTS successfully extracts object features and learns object relations from the segmentation network.
Advances in adversarial attacks and defenses in computer vision: A survey
In [2], we reviewed the contributions made by the computer vision community in adversarial attacks on deep learning (and their defenses) until the advent of year 2018.
Self-Supervised Video Object Segmentation by Motion-Aware Mask Propagation
We propose a self-supervised spatio-temporal matching method, coined Motion-Aware Mask Propagation (MAMP), for video object segmentation.
Controlled Caption Generation for Images Through Adversarial Attacks
In contrast, we propose a GAN-based algorithm for crafting adversarial examples for neural image captioning that mimics the internal representation of the CNN such that the resulting deep features of the input image enable a controlled incorrect caption generation through the recurrent network.
Attack to Fool and Explain Deep Networks
Exploring this phenomenon further, we alter the `adversarial' objective of our attack to use it as a tool to `explain' deep visual representation.
CAMERAS: Enhanced Resolution And Sanity preserving Class Activation Mapping for image saliency
Backpropagation image saliency aims at explaining model predictions by estimating model-centric importance of individual pixels in the input.
Survey: Image Mixing and Deleting for Data Augmentation
One such method is augmentation which introduces different types of corruption in the data to prevent the model from overfitting and to memorize patterns present in the data.
PC-DAN: Point Cloud based Deep Affinity Network for 3D Multi-Object Tracking (Accepted as an extended abstract in JRDB-ACT Workshop at CVPR21)
In recent times, the scope of LIDAR (Light Detection and Ranging) sensor-based technology has spread across numerous fields.
Physical world assistive signals for deep neural network classifiers -- neither defense nor attack
We discuss how we can exploit these insights to re-think, or avoid, some patterns that might contribute to, or degrade, the detectability of objects in the real-world.
Free-form Description Guided 3D Visual Graph Network for Object Grounding in Point Cloud
There are three main challenges in 3D object grounding: to find the main focus in the complex and diverse description; to understand the point cloud scene; and to locate the target object.
Picasso: A CUDA-based Library for Deep Learning over 3D Meshes
We present Picasso, a CUDA-based library comprising novel modules for deep learning over complex real-world 3D meshes.
LSDAT: Low-Rank and Sparse Decomposition for Decision-based Adversarial Attack
It is set forth that the proposed sparse perturbation is the most aligned sparse perturbation with the shortest path from the input sample to the decision boundary for some initial adversarial sample (the best sparse approximation of shortest path, likely to fool the model).
Deep Learning Based 3D Segmentation: A Survey
This paper fills the gap and provides a comprehensive survey of the recent progress made in deep learning based 3D segmentation.
Self-supervised Learning with Fully Convolutional Networks
By learning to solve a Jigsaw Puzzle problem with 25 patches and transferring the learned features to semantic segmentation task on Cityscapes dataset, we achieve a 5. 8 percentage point improvement over the baseline model that initialized from random values.
Defense-friendly Images in Adversarial Attacks: Dataset and Metrics for Perturbation Difficulty
We propose three metrics to determine the proportion of robust images in a dataset and provide scoring to determine the dataset bias.
Image Colorization: A Survey and Dataset
Image colorization is the process of estimating RGB colors for grayscale images or video frames to improve their aesthetic and perceptual quality.
Fast ORB-SLAM without Keypoint Descriptors
ORB-SLAM2 \cite{orbslam2} is a benchmark method in this domain, however, it consumes significant time for computing descriptors that never get reused unless a frame is selected as a keyframe.
Robotics Computational Geometry I.4.0; I.4.9
Simultaneous Detection and Tracking with Motion Modelling for Multiple Object Tracking
Deep learning-based Multiple Object Tracking (MOT) currently relies on off-the-shelf detectors for tracking-by-detection. This results in deep models that are detector biased and evaluations that are detector influenced.
Cassandra: Detecting Trojaned Networks from Adversarial Perturbations
We also propose an anomaly detection method to identify the target class in a Trojaned network.
Ranked #1 on
Adversarial Defense
on TrojAI Round 1
Odyssey: Creation, Analysis and Detection of Trojan Models
We propose a detector that is based on the analysis of the intrinsic DNN properties; that are affected due to the Trojaning process.
Orthogonal Deep Models As Defense Against Black-Box Attacks
On the other, deep learning has also been found vulnerable to adversarial attacks, which calls for new techniques to defend deep models against these attacks.
Minimum Potential Energy of Point Cloud for Robust Global Registration
Different from the most existing point set registration methods which usually extract the descriptors to find correspondences between point sets, our proposed MPE alignment method is able to handle large scale raw data offset without depending on traditional descriptors extraction, whether for the local or global registration methods.
Adversarial Perturbations Prevail in the Y-Channel of the YCbCr Color Space
Our results show that our approach achieves the best balance between defence against adversarial attacks such as FGSM, PGD and DDN and maintaining the original accuracies of VGG-16, ResNet50 and DenseNet121 on clean images.
Relation Graph Network for 3D Object Detection in Point Clouds
Convolutional Neural Networks (CNNs) have emerged as a powerful strategy for most object detection tasks on 2D images.
Empirical Autopsy of Deep Video Captioning Frameworks
We perform extensive experiments by varying the constituent components of the video captioning framework, and quantify the performance gains that are possible by mere component selection.
Deep localization of protein structures in fluorescence microscopy images
This study highlights the importance of deep learning for the analysis of fluorescence microscopy protein imagery.
Point Attention Network for Semantic Segmentation of 3D Point Clouds
We propose a point attention network that learns rich local shape features and their contextual correlations for 3D point cloud semantic segmentation.
Spherical Kernel for Efficient Graph Convolution on 3D Point Clouds
We propose a spherical kernel for efficient graph convolution of 3D point clouds.
Ranked #5 on
3D Object Classification
on ModelNet40
Adversarial Attack on Skeleton-based Human Action Recognition
We also explore the possibility of semantically imperceptible localized attacks with CIASA, and succeed in fooling the state-of-the-art skeleton action recognition models with high confidence.
Temporally Coherent Full 3D Mesh Human Pose Recovery from Monocular Video
A major challenge in this regard is the lack of appropriately annotated video data for learning the desired deep models.
Label Universal Targeted Attack
We introduce Label Universal Targeted Attack (LUTA) that makes a deep model predict a label of attacker's choice for `any' sample of a given source class with high probability.
Converting a Common Document Scanner to a Multispectral Scanner
We propose the construction of a prototype scanner designed to capture multispectral images of documents.
Multidimensional ground reaction forces and moments from wearable sensor accelerations via deep learning
Competing convolutional neural network (CNN) deep learning models were trained using laboratory-derived stance phase GRF/M data and simulated sensor accelerations for running and sidestepping maneuvers derived from nearly half a million legacy motion trials.
Octree guided CNN with Spherical Kernels for 3D Point Clouds
We propose an octree guided neural network architecture and spherical convolutional kernel for machine learning from arbitrary 3D point clouds.
Ranked #10 on
3D Part Segmentation
on ShapeNet-Part
Spatio-Temporal Dynamics and Semantic Attribute Enriched Visual Encoding for Video Captioning
The final representation is projected to a compact space and fed to a language model.
Deep Affinity Network for Multiple Object Tracking
In this paper, we harness the power of deep learning for data association in tracking by jointly modelling object appearances and their affinities between different frames in an end-to-end fashion.
On-field player workload exposure and knee injury risk monitoring via deep learning
In sports analytics, an understanding of accurate on-field 3D knee joint moments (KJM) could provide an early warning system for athlete workload exposure and knee injury risk.
Unsupervised Deep Multi-focus Image Fusion
Convolutional neural networks have recently been used for multi-focus image fusion.
Video Description: A Survey of Methods, Datasets and Evaluation Metrics
Video description is the automatic generation of natural language sentences that describe the contents of a given video.
Spherical Convolutional Neural Network for 3D Point Clouds
We propose a neural network for 3D point cloud processing that exploits `spherical' convolution kernels and octree partitioning of space.
Deep Keyframe Detection in Human Action Videos
Detecting representative frames in videos based on human actions is quite challenging because of the combined factors of human pose in action and the background.
Benchmark data and method for real-time people counting in cluttered scenes using depth sensors
A thorough evaluation on PCDS demonstrates that our technique is able to count people in cluttered scenes with high accuracy at 45 fps on a 1. 7 GHz processor, and hence it can be deployed for effective real-time people counting for intelligent transportation systems.
Hyperspectral recovery from RGB images using Gaussian Processes
We propose to recover spectral details from RGB images of known spectral quantization by modeling natural spectra under Gaussian Processes and combining them with the RGB images.
Threat of Adversarial Attacks on Deep Learning in Computer Vision: A Survey
This article presents the first comprehensive survey on adversarial attacks on deep learning in Computer Vision.
Defense against Universal Adversarial Perturbations
A rigorous evaluation shows that our framework can defend the network classifiers against unseen adversarial perturbations in the real-world scenarios with up to 97. 5% success rate.
3D Face Reconstruction from Light Field Images: A Model-free Approach
Reconstructing 3D facial geometry from a single RGB image has recently instigated wide research interest.
Learning from Millions of 3D Scans for Large-scale 3D Face Recognition
Unlike 2D photographs, 3D facial scans cannot be sourced from the web causing a bottleneck in the development of deep 3D face recognition networks and datasets.
Skepxels: Spatio-temporal Image Representation of Human Skeleton Joints for Action Recognition
The proposed action recognition exploits the representation in a hierarchical manner by first capturing the micro-temporal relations between the skeleton joints with the Skepxels and then exploiting their macro-temporal relations by computing the Fourier Temporal Pyramids over the CNN features of the skeletal images.
Viewpoint Invariant Action Recognition using RGB-D Videos
The proposed technique capitalizes on the spatio-temporal information available in the two data streams to the extract action features that are largely insensitive to the viewpoint variations.
Learning Human Pose Models from Synthesized Data for Robust RGB-D Action Recognition
We propose Human Pose Models that represent RGB and depth images of human poses independent of clothing textures, backgrounds, lighting conditions, body shapes and camera viewpoints.
Ranked #94 on
Skeleton Based Action Recognition
on NTU RGB+D
Modeling Sub-Event Dynamics in First-Person Action Recognition
We compare our method to state-of-the-art first person and generic video recognition algorithms.
Joint Discriminative Bayesian Dictionary and Classifier Learning
To further encourage discrimination in the dictionary, our model uses separate (sets of) Bernoulli distributions to represent data from different classes.
3D Action Recognition From Novel Viewpoints
We propose a human pose representation model that transfers human poses acquired from different unknown views to a view-invariant high-level space.
Learning a Deep Model for Human Action Recognition from Novel Viewpoints
The strength of our technique is that we learn a single R-NKTM for all actions and all viewpoints for knowledge transfer of any real human action video without the need for re-training or fine-tuning the model.
Sparseness helps: Sparsity Augmented Collaborative Representation for Classification
Many classification approaches first represent a test sample using the training samples of all the classes.
Shape-Based Automatic Detection of a Large Number of 3D Facial Landmarks
Our approach does not use texture and is completely shape based in order to detect landmarks that are morphologically significant.
Bayesian Sparse Representation for Hyperspectral Image Super Resolution
We propose a hyperspectral image super resolution approach that fuses a high resolution image with the low resolution hyperspectral image using non-parametric Bayesian sparse representation.
Learning a Non-Linear Knowledge Transfer Model for Cross-View Action Recognition
We propose unsupervised learning of a non-linear model that transfers knowledge from multiple views to a canonical view.
Discriminative Bayesian Dictionary Learning for Classification
We propose a Bayesian approach to learn discriminative dictionaries for sparse representation of data.
Representation Learning with Deep Extreme Learning Machines for Efficient Image Set Classification
Efficient and accurate joint representation of a collection of images, that belong to the same class, is a major research challenge for practical image set classification.
Dense 3D Face Correspondence
A deformable model (K3DM) is constructed from the dense corresponded faces and an algorithm is proposed for morphing the K3DM to fit unseen faces.
Histogram of Oriented Principal Components for Cross-View Action Recognition
We propose the Histogram of Oriented Principal Components (HOPC) descriptor that is robust to noise, viewpoint, scale and action speed variations.
HOPC: Histogram of Oriented Principal Components of 3D Pointclouds for Action Recognition
In contrast, we directly process the pointclouds and propose a new technique for action recognition which is more robust to noise, action speed and viewpoint variations.
Action Classification with Locality-constrained Linear Coding
We use the Histogram of Oriented Gradient (HOG3D) feature to encode the information in each cell.
Optimizing Auto-correlation for Fast Target Search in Large Search Space
To this end we propose an Efficient Group Size (EGS) algorithm which minimizes the number of similarity computations for a particular search image.
Semi-supervised Spectral Clustering for Image Set Classification
We present an image set classification algorithm based on unsupervised clustering of labeled training and unlabeled test data where labels are only used in the stopping criterion.
Multispectral Palmprint Encoding and Recognition
Comprehensive experiments for both identification and verification scenarios are performed on two public datasets -- one captured with a contact-based sensor (PolyU dataset), and the other with a contact-free sensor (CASIA dataset).
