Search Results for author:
Found 79 papers, 24 papers with code
On Modulating the Gradient for Meta-Learning
Inspired by optimization techniques, we propose a novel meta-learning algorithm with gradient modulation to encourage fast-adaptation of neural networks in the absence of abundant data.
CHAIN: Enhancing Generalization in Data-Efficient GANs via lipsCHitz continuity constrAIned Normalization
Our work addresses this gap by identifying a critical flaw in BN: the tendency for gradient explosion during the centering and scaling steps.
Learning Gaussian Representation for Eye Fixation Prediction
Existing eye fixation prediction methods perform the mapping from input images to the corresponding dense fixation maps generated from raw fixation points.
Meet JEANIE: a Similarity Measure for 3D Skeleton Sequences via Temporal-Viewpoint Alignment
For a support sequence, we match it with view-simulated query sequences, as in the popular Dynamic Time Warping (DTW).
Few-shot Shape Recognition by Learning Deep Shape-aware Features
Thirdly, we propose a decoding module to include the supervision of shape masks and edges and align the original and reconstructed shape features, enforcing the learned features to be more shape-aware.
Pre-training with Random Orthogonal Projection Image Modeling
In this paper, we propose an Image Modeling framework based on random orthogonal projection instead of binary masking as in MIM.
Alignment and Outer Shell Isotropy for Hyperbolic Graph Contrastive Learning
We show that in the hyperbolic space one has to address the leaf- and height-level uniformity which are related to properties of trees, whereas in the ambient space of the hyperbolic manifold, these notions translate into imposing an isotropic ring density towards boundaries of Poincar\'e ball.
Flow Dynamics Correction for Action Recognition
Various research studies indicate that action recognition performance highly depends on the types of motions being extracted and how accurate the human actions are represented.
Adaptive Multi-head Contrastive Learning
As such, enforcing a high similarity for positive pairs and a low similarity for negative pairs may not always be achievable, and in the case of some pairs, forcing so may be detrimental to the performance.
Multivariate Prototype Representation for Domain-Generalized Incremental Learning
Deep learning models suffer from catastrophic forgetting when being fine-tuned with samples of new classes.
Exploiting Field Dependencies for Learning on Categorical Data
Instead of modelling statistics of features globally (i. e., by the covariance matrix of features), we learn a global field dependency matrix that captures dependencies between fields and then we refine the global field dependency matrix at the instance-wise level with different weights (so-called local dependency modelling) w. r. t.
Distilling Self-Supervised Vision Transformers for Weakly-Supervised Few-Shot Classification & Segmentation
For this mixed setup, we propose to improve the pseudo-labels using a pseudo-label enhancer that was trained using the available ground-truth pixel-level labels.
Traffic Forecasting on New Roads Using Spatial Contrastive Pre-Training (SCPT)
During inference, the spatial encoder only requires two days of traffic data on the new roads and does not require any re-training.
Message Passing Neural Networks for Traffic Forecasting
A road network, in the context of traffic forecasting, is typically modeled as a graph where the nodes are sensors that measure traffic metrics (such as speed) at that location.
Transductive Few-shot Learning with Prototype-based Label Propagation by Iterative Graph Refinement
In this paper, we propose a novel prototype-based label propagation to solve these issues.
Learning Partial Correlation based Deep Visual Representation for Image Classification
Our work obtains a partial correlation based deep visual representation and mitigates the small sample problem often encountered by covariance matrix estimation in CNN.
From Saliency to DINO: Saliency-guided Vision Transformer for Few-shot Keypoint Detection
Unlike current deep keypoint detectors that are trained to recognize limited number of body parts, few-shot keypoint detection (FSKD) attempts to localize any keypoints, including novel or base keypoints, depending on the reference samples.
3Mformer: Multi-order Multi-mode Transformer for Skeletal Action Recognition
We split action sequences into temporal blocks, Higher-order Transformer (HoT) produces embeddings of each temporal block based on (i) the body joints, (ii) pairwise links of body joints and (iii) higher-order hyper-edges of skeleton body joints.
Because Every Sensor Is Unique, so Is Every Pair: Handling Dynamicity in Traffic Forecasting
Further analysis also shows that each pair of sensors also has a unique dynamic.
Event-guided Multi-patch Network with Self-supervision for Non-uniform Motion Deblurring
Contemporary deep learning multi-scale deblurring models suffer from many issues: 1) They perform poorly on non-uniformly blurred images/videos; 2) Simply increasing the model depth with finer-scale levels cannot improve deblurring; 3) Individual RGB frames contain a limited motion information for deblurring; 4) Previous models have a limited robustness to spatial transformations and noise.
Learning Spatial-context-aware Global Visual Feature Representation for Instance Image Retrieval
To this end, we propose a novel feature learning framework for instance image retrieval, which embeds local spatial context information into the learned global feature representations.
Spectral Feature Augmentation for Graph Contrastive Learning and Beyond
Although augmentations (e. g., perturbation of graph edges, image crops) boost the efficiency of Contrastive Learning (CL), feature level augmentation is another plausible, complementary yet not well researched strategy.
Temporal-Viewpoint Transportation Plan for Skeletal Few-shot Action Recognition
To factor out misalignment between query and support sequences of 3D body joints, we propose an advanced variant of Dynamic Time Warping which jointly models each smooth path between the query and support frames to achieve simultaneously the best alignment in the temporal and simulated camera viewpoint spaces for end-to-end learning under the limited few-shot training data.
Uncertainty-DTW for Time Series and Sequences
Dynamic Time Warping (DTW) is used for matching pairs of sequences and celebrated in applications such as forecasting the evolution of time series, clustering time series or even matching sequence pairs in few-shot action recognition.
Time-rEversed diffusioN tEnsor Transformer: A new TENET of Few-Shot Object Detection
To address these drawbacks, we propose a Time-rEversed diffusioN tEnsor Transformer (TENET), which i) forms high-order tensor representations that capture multi-way feature occurrences that are highly discriminative, and ii) uses a transformer that dynamically extracts correlations between the query image and the entire support set, instead of a single average-pooled support embedding.
COSTA: Covariance-Preserving Feature Augmentation for Graph Contrastive Learning
In this paper, we show that the node embedding obtained via the graph augmentations is highly biased, somewhat limiting contrastive models from learning discriminative features for downstream tasks.
Accurate 3-DoF Camera Geo-Localization via Ground-to-Satellite Image Matching
We address the problem of ground-to-satellite image geo-localization, that is, estimating the camera latitude, longitude and orientation (azimuth angle) by matching a query image captured at the ground level against a large-scale database with geotagged satellite images.
Graph-adaptive Rectified Linear Unit for Graph Neural Networks
However, in the updating stage, all nodes share the same updating function.
Multi-level Second-order Few-shot Learning
The goal of multi-level feature design is to extract feature representations at different layer-wise levels of CNN, realizing several levels of visual abstraction to achieve robust few-shot learning.
Contrastive Laplacian Eigenmaps
Starting from a GAN-inspired contrastive formulation, we show that the Jensen-Shannon divergence underlying many contrastive graph embedding models fails under disjoint positive and negative distributions, which may naturally emerge during sampling in the contrastive setting.
Kernelized Few-Shot Object Detection With Efficient Integral Aggregation
We design a Kernelized Few-shot Object Detector by leveraging kernelized matrices computed over multiple proposal regions, which yield expressive non-linear representations whose model complexity is learned on the fly.
EASE: Unsupervised Discriminant Subspace Learning for Transductive Few-Shot Learning
We present an unsupErvised discriminAnt Subspace lEarning (EASE) that improves transductive few-shot learning performance by learning a linear projection onto a subspace built from features of the support set and the unlabeled query set in the test time.
3D Skeleton-based Few-shot Action Recognition with JEANIE is not so Naïve
In this paper, we propose a Few-shot Learning pipeline for 3D skeleton-based action recognition by Joint tEmporal and cAmera viewpoiNt alIgnmEnt (JEANIE).
Manifold Learning Benefits GANs
In our design, the manifold learning and coding steps are intertwined with layers of the discriminator, with the goal of attracting intermediate feature representations onto manifolds.
Few-shot Keypoint Detection with Uncertainty Learning for Unseen Species
Current non-rigid object keypoint detectors perform well on a chosen kind of species and body parts, and require a large amount of labelled keypoints for training.
Meta-Learning for Multi-Label Few-Shot Classification
Even with the luxury of having abundant data, multi-label classification is widely known to be a challenging task to address.
Towards a Robust Differentiable Architecture Search under Label Noise
Our empirical evaluations show that the noise injecting operation does not degrade the performance of the NAS algorithm if the data is indeed clean.
Simple Dialogue System with AUDITED
For the image-based task, we employ the DeepFashion dataset in which we seek nearest neighbor images of positive and negative target images of the MMD data.
High-order Tensor Pooling with Attention for Action Recognition
We aim at capturing high-order statistics of feature vectors formed by a neural network, and propose end-to-end second- and higher-order pooling to form a tensor descriptor.
Ranked #2 on
Scene Recognition
on YUP++
(using extra training data)
REFINE: Random RangE FInder for Network Embedding
Moreover, we design a simple but efficient spectral filter for network enhancement to obtain higher-order information for node representation.
Graph Convolutional Network with Generalized Factorized Bilinear Aggregation
Although Graph Convolutional Networks (GCNs) have demonstrated their power in various applications, the graph convolutional layers, as the most important component of GCN, are still using linear transformations and a simple pooling step.
Predicting Flight Delay with Spatio-Temporal Trajectory Convolutional Network and Airport Situational Awareness Map
To model and forecast flight delays accurately, it is crucial to harness various vehicle trajectory and contextual sensor data on airport tarmac areas.
On Learning the Geodesic Path for Incremental Learning
This in practice is done by minimizing the dissimilarity between current and previous responses of the network one way or another.
Simple Spectral Graph Convolution
Our spectral analysis shows that our simple spectral graph convolution used in S^2GC is a low-pass filter which partitions networks into a few large parts.
Ranked #1 on
Node Clustering
on Wiki
Tensor Representations for Action Recognition
In this paper, we propose novel tensor representations for compactly capturing such higher-order relationships between visual features for the task of action recognition.
Ranked #2 on
Skeleton Based Action Recognition
on UT-Kinect
Action Recognition In Videos
Skeleton Based Action Recognition
Power Normalizations in Fine-grained Image, Few-shot Image and Graph Classification
Our layer combines the feature vectors and their respective spatial locations in the feature maps produced by the last convolutional layer of CNN into a positive definite matrix with second-order statistics to which PN operators are applied, forming so-called Second-order Pooling (SOP).
A Token-wise CNN-based Method for Sentence Compression
Sentence compression is a Natural Language Processing (NLP) task aimed at shortening original sentences and preserving their key information.
6DoF Object Pose Estimation via Differentiable Proxy Voting Loss
In this paper, we aim to reduce such errors by incorporating the distances between pixels and keypoints into our objective.
Relation Embedding for Personalised POI Recommendation
Point-of-Interest (POI) recommendation is one of the most important location-based services helping people discover interesting venues or services.
Self-supervising Action Recognition by Statistical Moment and Subspace Descriptors
In this paper, we build on a concept of self-supervision by taking RGB frames as input to learn to predict both action concepts and auxiliary descriptors e. g., object descriptors.
Ranked #1 on
Scene Recognition
on YUP++
(using extra training data)
Rethinking Class Relations: Absolute-relative Supervised and Unsupervised Few-shot Learning
The majority of existing few-shot learning methods describe image relations with binary labels.
Few-shot Action Recognition with Permutation-invariant Attention
Such encoded blocks are aggregated by permutation-invariant pooling to make our approach robust to varying action lengths and long-range temporal dependencies whose patterns are unlikely to repeat even in clips of the same class.
Ranked #6 on
Few Shot Action Recognition
on Kinetics-100
Improving Few-shot Learning by Spatially-aware Matching and CrossTransformer
In this paper, we study the impact of scale and location mismatch in the few-shot learning scenario, and propose a novel Spatially-aware Matching (SM) scheme to effectively perform matching across multiple scales and locations, and learn image relations by giving the highest weights to the best matching pairs.
COLTRANE: ConvolutiOnaL TRAjectory NEtwork for Deep Map Inference
This paper introduces COLTRANE, ConvolutiOnaL TRAjectory NEtwork, a novel deep map inference framework which operates on GPS trajectories collected in various environments.
A Comparative Review of Recent Kinect-based Action Recognition Algorithms
Video-based human action recognition is currently one of the most active research areas in computer vision.
Ranked #85 on
Skeleton Based Action Recognition
on NTU RGB+D
Hallucinating IDT Descriptors and I3D Optical Flow Features for Action Recognition with CNNs
Thus, we propose an end-to-end trainable network with streams which learn the IDT-based BoW/FV representations at the training stage and are simple to integrate with the I3D model.
Ranked #3 on
Scene Recognition
on YUP++
(using extra training data)
Projective Subspace Networks For Few-Shot Learning
Generalization from limited examples, usually studied under the umbrella of meta-learning, equips learning techniques with the ability to adapt quickly in dynamical environments and proves to be an essential aspect of lifelong learning.
Recovering Faces from Portraits with Auxiliary Facial Attributes
%Our method can recover high-quality photorealistic faces from unaligned portraits while preserving the identity of the face images as well as it can reconstruct a photorealistic face image with a desired set of attributes.
Identity-preserving Face Recovery from Stylized Portraits
We develop an Identity-preserving Face Recovery from Portraits (IFRP) method that utilizes a Style Removal network (SRN) and a Discriminative Network (DN).
Deep Stacked Hierarchical Multi-patch Network for Image Deblurring
depth, we propose a stacked version of our multi-patch model.
Ranked #9 on
Deblurring
on RealBlur-R (trained on GoPro)
(SSIM (sRGB) metric)
Few-Shot Learning via Saliency-guided Hallucination of Samples
To the best of our knowledge, we are the first to leverage saliency maps for such a task and we demonstrate their usefulness in hallucinating additional datapoints for few-shot learning.
Fisher-Bures Adversary Graph Convolutional Networks
We try to minimize the loss wrt the perturbed $G+\Delta{G}$ while making $\Delta{G}$ to be effective in terms of the Fisher information of the neural network.
Power Normalizing Second-order Similarity Network for Few-shot Learning
In this paper, we propose a similarity learning network leveraging second-order information and Power Normalizations.
Model Selection for Generalized Zero-shot Learning
Specifically, we leverage two sources of datapoints (observed and auxiliary) to train some classifier to recognize which test datapoints come from seen and which from unseen classes.
Generalized Zero-Shot Learning
Generative Adversarial Network
+1
Museum Exhibit Identification Challenge for the Supervised Domain Adaptation and Beyond
To achieve robust baselines, we build on a recent approach that aligns per-class scatter matrices of the source and target CNN streams.
Second-order Democratic Aggregation
In this paper, we study a class of orderless aggregation functions designed to minimize interference or equalize contributions in the context of second-order features and we show that they can be computed just as efficiently as their first-order counterparts and they have favorable properties over aggregation by summation.
A Deeper Look at Power Normalizations
In this paper, we reconsider these operators in the deep learning setup by introducing a novel layer that implements PN for non-linear pooling of feature maps.
CNN-based Action Recognition and Supervised Domain Adaptation on 3D Body Skeletons via Kernel Feature Maps
In this paper, we propose a new representation which encodes sequences of 3D body skeleton joints in texture-like representations derived from mathematically rigorous kernel methods.
Artwork Identification from Wearable Camera Images for Enhancing Experience of Museum Audiences
We discuss the application of wearable cameras, and the practical and technical challenges in devising a robust system that can recognize artworks viewed by the visitors to create a detailed record of their visit.
Zero-Shot Kernel Learning
In contrast, we apply well-established kernel methods to learn a non-linear mapping between the feature and attribute spaces.
Face Destylization
To enforce the destylized faces to be similar to authentic face images, we employ a discriminative network, which consists of convolutional and fully connected layers.
Museum Exhibit Identification Challenge for Domain Adaptation and Beyond
To achieve robust baselines, we build on a recent approach that aligns per-class scatter matrices of the source and target CNN streams [15].
Identity-preserving Face Recovery from Portraits
In this paper, we present a new Identity-preserving Face Recovery from Portraits (IFRP) to recover latent photorealistic faces from unaligned stylized portraits.
Higher-order Pooling of CNN Features via Kernel Linearization for Action Recognition
The HOK descriptors are then generated from the higher-order co-occurrences of these feature maps, and are then used as input to a video-level classifier.
Domain Adaptation by Mixture of Alignments of Second- or Higher-Order Scatter Tensors
In this paper, we propose an approach to the domain adaptation, dubbed Second- or Higher-order Transfer of Knowledge (So-HoT), based on the mixture of alignments of second- or higher-order scatter statistics between the source and target domains.
Sparse Coding for Third-Order Super-Symmetric Tensor Descriptors With Application to Texture Recognition
Super-symmetric tensors - a higher-order extension of scatter matrices - are becoming increasingly popular in machine learning and computer vision for modeling data statistics, co-occurrences, or even as visual descriptors.
Tensor Representations via Kernel Linearization for Action Recognition from 3D Skeletons (Extended Version)
We first define RBF kernels on 3D joint sequences, which are then linearized to form kernel descriptors.
Dictionary Learning and Sparse Coding for Third-order Super-symmetric Tensors
Super-symmetric tensors - a higher-order extension of scatter matrices - are becoming increasingly popular in machine learning and computer vision for modelling data statistics, co-occurrences, or even as visual descriptors.
Convolutional Kernel Networks
An important goal in visual recognition is to devise image representations that are invariant to particular transformations.
Ranked #23 on
Image Classification
on MNIST
