Search Results for author:
Found 111 papers, 23 papers with code
CLNet: A Compact Latent Network for Fast Adjusting Siamese Trackers
To make full use of these sequence-specific samples, {we propose a compact latent network to quickly adjust the tracking model to adapt to new scenes.}
QuickSRNet: Plain Single-Image Super-Resolution Architecture for Faster Inference on Mobile Platforms
In this work, we present QuickSRNet, an efficient super-resolution architecture for real-time applications on mobile platforms.
X$^3$KD: Knowledge Distillation Across Modalities, Tasks and Stages for Multi-Camera 3D Object Detection
Specifically, we propose cross-task distillation from an instance segmentation teacher (X-IS) in the PV feature extraction stage providing supervision without ambiguous error backpropagation through the view transformation.
DejaVu: Conditional Regenerative Learning to Enhance Dense Prediction
Next, we use a conditional regenerator, which takes the redacted image and the dense predictions as inputs, and reconstructs the original image by filling in the missing structural information.
TransAdapt: A Transformative Framework for Online Test Time Adaptive Semantic Segmentation
Test-time adaptive (TTA) semantic segmentation adapts a source pre-trained image semantic segmentation model to unlabeled batches of target domain test images, different from real-world, where samples arrive one-by-one in an online fashion.
Adaptive Siamese Tracking with a Compact Latent Network
Under this viewing, we perform an in-depth analysis for them through visual simulations and real tracking examples, and find that the failure cases in some challenging situations can be regarded as the issue of missing decisive samples in offline training.
Skip-Attention: Improving Vision Transformers by Paying Less Attention
This work aims to improve the efficiency of vision transformers (ViT).
Guidance Through Surrogate: Towards a Generic Diagnostic Attack
Recently, different adversarial training defenses are proposed that not only maintain a high clean accuracy but also show significant robustness against popular and well studied adversarial attacks such as PGD.
Test-time Adaptation vs. Training-time Generalization: A Case Study in Human Instance Segmentation using Keypoints Estimation
In this approach, we do not assume test-time access to the labeled source dataset.
PartSLIP: Low-Shot Part Segmentation for 3D Point Clouds via Pretrained Image-Language Models
Generalizable 3D part segmentation is important but challenging in vision and robotics.
Self-Supervised Geometric Correspondence for Category-Level 6D Object Pose Estimation in the Wild
While 6D object pose estimation has wide applications across computer vision and robotics, it remains far from being solved due to the lack of annotations.
X-Align: Cross-Modal Cross-View Alignment for Bird's-Eye-View Segmentation
Bird's-eye-view (BEV) grid is a common representation for the perception of road components, e. g., drivable area, in autonomous driving.
Online Adaptive Personalization for Face Anti-spoofing
Face authentication systems require a robust anti-spoofing module as they can be deceived by fabricating spoof images of authorized users.
Learning Implicit Feature Alignment Function for Semantic Segmentation
As such, IFA implicitly aligns the feature maps at different levels and is capable of producing segmentation maps in arbitrary resolutions.
IRISformer: Dense Vision Transformers for Single-Image Inverse Rendering in Indoor Scenes
Indoor scenes exhibit significant appearance variations due to myriad interactions between arbitrarily diverse object shapes, spatially-changing materials, and complex lighting.
Simple and Efficient Architectures for Semantic Segmentation
Though the state-of-the architectures for semantic segmentation, such as HRNet, demonstrate impressive accuracy, the complexity arising from their salient design choices hinders a range of model acceleration tools, and further they make use of operations that are inefficient on current hardware.
Imposing Consistency for Optical Flow Estimation
Imposing consistency through proxy tasks has been shown to enhance data-driven learning and enable self-supervision in various tasks.
Panoptic, Instance and Semantic Relations: A Relational Context Encoder to Enhance Panoptic Segmentation
A Panoptic Relational Attention (PRA) module is then applied to the encodings and the global feature map from the backbone.
SALISA: Saliency-based Input Sampling for Efficient Video Object Detection
In this paper, we propose SALISA, a novel non-uniform SALiency-based Input SAmpling technique for video object detection that allows for heavy down-sampling of unimportant background regions while preserving the fine-grained details of a high-resolution image.
Delta Distillation for Efficient Video Processing
By extensive experiments on a wide range of architectures, including the most efficient ones, we demonstrate that delta distillation sets a new state of the art in terms of accuracy vs. efficiency trade-off for semantic segmentation and object detection in videos.
Ranked #2 on
Video Semantic Segmentation
on Cityscapes val
Consistency and Diversity induced Human Motion Segmentation
Besides, a novel constraint based on the Hilbert Schmidt Independence Criterion (HSIC) is introduced to ensure the diversity of multi-level subspace representations, which enables the complementarity of multi-level representations to be explored to boost the transfer learning performance.
Real-Time, Accurate, and Consistent Video Semantic Segmentation via Unsupervised Adaptation and Cross-Unit Deployment on Mobile Device
Such a deployment scheme best utilizes the available processing power on the smartphone and enables real-time operation of our adaptive video segmentation algorithm.
Modality-Agnostic Topology Aware Localization
This work presents a data-driven approach for the indoor localization of an observer on a 2D topological map of the environment.
Distribution Estimation to Automate Transformation Policies for Self-Supervision
However, it is observed that image transformations already present in the dataset might be less effective in learning such self-supervised representations.
Dynamic Iterative Refinement for Efficient 3D Hand Pose Estimation
While hand pose estimation is a critical component of most interactive extended reality and gesture recognition systems, contemporary approaches are not optimized for computational and memory efficiency.
HS3: Learning with Proper Task Complexity in Hierarchically Supervised Semantic Segmentation
While deeply supervised networks are common in recent literature, they typically impose the same learning objective on all transitional layers despite their varying representation powers.
Ranked #10 on
Semantic Segmentation
on NYU Depth v2
X-Distill: Improving Self-Supervised Monocular Depth via Cross-Task Distillation
In order to enable such knowledge distillation across two different visual tasks, we introduce a small, trainable network that translates the predicted depth map to a semantic segmentation map, which can then be supervised by the teacher network.
AuxAdapt: Stable and Efficient Test-Time Adaptation for Temporally Consistent Video Semantic Segmentation
Since inconsistency mainly arises from the model's uncertainty in its output, we propose an adaptation scheme where the model learns from its own segmentation decisions as it streams a video, which allows producing more confident and temporally consistent labeling for similarly-looking pixels across frames.
Perceptual Consistency in Video Segmentation
More specifically, by measuring the perceptual consistency between the predicted segmentation and the available ground truth on a nearby frame and combining it with the segmentation confidence, we can accurately assess the classification correctness on each pixel.
ConFeSS: A Framework for Single Source Cross-Domain Few-Shot Learning
The first step of our framework trains a feature extracting backbone with the contrastive loss on the base category data.
A Survey on Deep Learning Technique for Video Segmentation
Video segmentation -- partitioning video frames into multiple segments or objects -- plays a critical role in a broad range of practical applications, from enhancing visual effects in movie, to understanding scenes in autonomous driving, to creating virtual background in video conferencing.
On Improving Adversarial Transferability of Vision Transformers
(ii) Token Refinement: We then propose to refine the tokens to further enhance the discriminative capacity at each block of ViT.
Data-driven Weight Initialization with Sylvester Solvers
The initialization is cast as an optimization problem where we minimize a combination of encoding and decoding losses of the input activations, which is further constrained by a user-defined latent code.
InverseForm: A Loss Function for Structured Boundary-Aware Segmentation
We present a novel boundary-aware loss term for semantic segmentation using an inverse-transformation network, which efficiently learns the degree of parametric transformations between estimated and target boundaries.
Ranked #14 on
Semantic Segmentation
on NYU Depth v2
On Generating Transferable Targeted Perturbations
To this end, we propose a new objective function that not only aligns the global distributions of source and target images, but also matches the local neighbourhood structure between the two domains.
Cascade Weight Shedding in Deep Neural Networks: Benefits and Pitfalls for Network Pruning
We report, for the first time, on the cascade weight shedding phenomenon in deep neural networks where in response to pruning a small percentage of a network's weights, a large percentage of the remaining is shed over a few epochs during the ensuing fine-tuning phase.
Efficient Action Recognition via Dynamic Knowledge Propagation
Efficient action recognition has become crucial to extend the success of action recognition to many real-world applications.
Structured Convolutions for Efficient Neural Network Design
We show how this decomposition can be applied to 2D and 3D kernels as well as the fully-connected layers.
Stylized Adversarial Defense
In contrast to existing adversarial training methods that only use class-boundary information (e. g., using a cross-entropy loss), we propose to exploit additional information from the feature space to craft stronger adversaries that are in turn used to learn a robust model.
Towards Purely Unsupervised Disentanglement of Appearance and Shape for Person Images Generation
To be clear, in this paper, we refer unsupervised learning as learning without task-specific human annotations, pairs or any form of weak supervision.)
A Self-supervised Approach for Adversarial Robustness
Adversarial examples can cause catastrophic mistakes in Deep Neural Network (DNNs) based vision systems e. g., for classification, segmentation and object detection.
Identity Enhanced Residual Image Denoising
We propose to learn a fully-convolutional network model that consists of a Chain of Identity Mapping Modules and residual on the residual architecture for image denoising.
See More, Know More: Unsupervised Video Object Segmentation with Co-Attention Siamese Networks
We introduce a novel network, called CO-attention Siamese Network (COSNet), to address the unsupervised video object segmentation task from a holistic view.
Semantic Segmentation
Unsupervised Video Object Segmentation
+2
Image Segmentation Using Deep Learning: A Survey
Image segmentation is a key topic in image processing and computer vision with applications such as scene understanding, medical image analysis, robotic perception, video surveillance, augmented reality, and image compression, among many others.
Sparse Coding on Cascaded Residuals
The residual at a layer is computed by the difference between the aggregated reconstructions of the previous layers and the downsampled original image at that layer.
Component Attention Guided Face Super-Resolution Network: CAGFace
To make the best use of the underlying structure of faces, the collective information through face datasets and the intermediate estimates during the upsampling process, here we introduce a fully convolutional multi-stage neural network for 4$\times$ super-resolution for face images.
Sketch-Specific Data Augmentation for Freehand Sketch Recognition
Conventional methods for this task often rely on the availability of the temporal order of sketch strokes, additional cues acquired from different modalities and supervised augmentation of sketch datasets with real images, which also limit the applicability and feasibility of these methods in real scenarios.
Joint Stereo Video Deblurring, Scene Flow Estimation and Moving Object Segmentation
Under our model, these three tasks are naturally connected and expressed as the parameter estimation of 3D scene structure and camera motion (structure and motion for the dynamic scenes).
Deep Ancient Roman Republican Coin Classification via Feature Fusion and Attention
We perform the classification of ancient Roman Republican coins via recognizing their reverse motifs where various objects, faces, scenes, animals, and buildings are minted along with legends.
Towards better Validity: Dispersion based Clustering for Unsupervised Person Re-identification
With this insight, we design a novel Dispersion-based Clustering (DBC) approach which can discover the underlying patterns in data.
Ranked #13 on
Unsupervised Person Re-Identification
on Market-1501
Cross-Domain Transferability of Adversarial Perturbations
To this end, we propose a framework capable of launching highly transferable attacks that crafts adversarial patterns to mislead networks trained on wholly different domains.
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).
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.
Task-generalizable Adversarial Attack based on Perceptual Metric
Deep neural networks (DNNs) can be easily fooled by adding human imperceptible perturbations to the images.
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.
Face Super-resolution Guided by Facial Component Heatmaps
State-of-the-art face super-resolution methods use deep convolutional neural networks to learn a mapping between low-resolution (LR) facial patterns and their corresponding high-resolution (HR) counterparts by exploring local information.
Deep Underwater Image Enhancement
In an underwater scene, wavelength-dependent light absorption and scattering degrade the visibility of images, causing low contrast and distorted color casts.
Local Gradients Smoothing: Defense against localized adversarial attacks
Deep neural networks (DNNs) have shown vulnerability to adversarial attacks, i. e., carefully perturbed inputs designed to mislead the network at inference time.
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.
Hyperparameter Optimization for Tracking With Continuous Deep Q-Learning
Hyperparameters are numerical presets whose values are assigned prior to the commencement of the learning process.
One-Shot Action Localization by Learning Sequence Matching Network
Learning based temporal action localization methods require vast amounts of training data.
Super-Resolving Very Low-Resolution Face Images With Supplementary Attributes
An LR input contains low-frequency facial components of its HR version while its residual face image defined as the difference between the HR ground-truth and interpolated LR images contains the missing high-frequency facial details.
Feature Affinity based Pseudo Labeling for Semi-supervised Person Re-identification
Our approach measures the affinity of unlabeled samples with the underlying clusters of labeled data samples using the intermediate feature representations from deep networks.
Video Representation Learning Using Discriminative Pooling
In an attempt to tackle this problem, we propose discriminative pooling, based on the notion that among the deep features generated on all short clips, there is at least one that characterizes the action.
A Cascaded Convolutional Neural Network for Single Image Dehazing
Different from previous learning-based methods, we propose a flexible cascaded CNN for single hazy image restoration, which considers the medium transmission and global atmospheric light jointly by two task-driven subnetworks.
Zero-Shot Object Detection: Learning to Simultaneously Recognize and Localize Novel Concepts
We hypothesize that this setting is ill-suited for real-world applications where unseen objects appear only as a part of a complex scene, warranting both the `recognition' and `localization' of an unseen category.
Indoor Scene Understanding in 2.5/3D for Autonomous Agents: A Survey
With the availability of low-cost and compact 2. 5/3D visual sensing devices, computer vision community is experiencing a growing interest in visual scene understanding of indoor environments.
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.
Chaining Identity Mapping Modules for Image Denoising
We propose to learn a fully-convolutional network model that consists of a Chain of Identity Mapping Modules (CIMM) for image denoising.
DR-Net: Transmission Steered Single Image Dehazing Network with Weakly Supervised Refinement
Despite the recent progress in image dehazing, several problems remain largely unsolved such as robustness for varying scenes, the visual quality of reconstructed images, and effectiveness and flexibility for applications.
Depth Map Completion by Jointly Exploiting Blurry Color Images and Sparse Depth Maps
In this paper, we propose to tackle the problem of depth map completion by jointly exploiting the blurry color image sequences and the sparse depth map measurements, and present an energy minimization based formulation to simultaneously complete the depth maps, estimate the scene flow and deblur the color images.
Regularization of Deep Neural Networks with Spectral Dropout
We cast the proposed approach in the form of regular Convolutional Neural Network (CNN) weight layers using a decorrelation transform with fixed basis functions.
Let Features Decide for Themselves: Feature Mask Network for Person Re-identification
Person re-identification aims at establishing the identity of a pedestrian from a gallery that contains images of multiple people obtained from a multi-camera system.
Multipartite Pooling for Deep Convolutional Neural Networks
We propose a novel pooling strategy that learns how to adaptively rank deep convolutional features for selecting more informative representations.
Scene Categorization With Spectral Features
To the best of our knowledge, this is the first attempt to use deep learning based spectral features explicitly for image classification task.
Deep Edge-Aware Saliency Detection
There has been profound progress in visual saliency thanks to the deep learning architectures, however, there still exist three major challenges that hinder the detection performance for scenes with complex compositions, multiple salient objects, and salient objects of diverse scales.
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.
Hallucinating Very Low-Resolution Unaligned and Noisy Face Images by Transformative Discriminative Autoencoders
Then we use a transformative encoder network to project the intermediate HR faces to aligned and noise-free LR faces.
Ranked #7 on
Image Super-Resolution
on VggFace2 - 8x upscaling
A Unified approach for Conventional Zero-shot, Generalized Zero-shot and Few-shot Learning
Then, it learns how to combine these directions to obtain the principal direction for each unseen class such that the CAPD of the test image is aligned with the semantic embedding of the true class, and opposite to the other classes.
Integrated Deep and Shallow Networks for Salient Object Detection
Then the results from deep FCNN and RBD are concatenated to feed into a shallow network to map the concatenated feature maps to saliency maps.
Quadruplet Network with One-Shot Learning for Fast Visual Object Tracking
In this paper, we propose a new quadruplet deep network to examine the potential connections among the training instances, aiming to achieve a more powerful representation.
Simultaneous Stereo Video Deblurring and Scene Flow Estimation
Unlike the existing approach [31] which used a pre-computed scene flow, we propose a single framework to jointly estimate the scene flow and deblur the image, where the motion cues from scene flow estimation and blur information could reinforce each other, and produce superior results than the conventional scene flow estimation or stereo deblurring methods.
Action Representation Using Classifier Decision Boundaries
Applying multiple instance learning in an SVM setup, we use the parameters of this separating hyperplane as a descriptor for the video.
Selective Video Object Cutout
Conventional video segmentation approaches rely heavily on appearance models.
Super-Trajectory for Video Segmentation
We introduce a novel semi-supervised video segmentation approach based on an efficient video representation, called as "super-trajectory".
Ordered Pooling of Optical Flow Sequences for Action Recognition
Training of Convolutional Neural Networks (CNNs) on long video sequences is computationally expensive due to the substantial memory requirements and the massive number of parameters that deep architectures demand.
Robust and Real-time Deep Tracking Via Multi-Scale Domain Adaptation
Visual tracking is a fundamental problem in computer vision.
Learning an Invariant Hilbert Space for Domain Adaptation
This paper introduces a learning scheme to construct a Hilbert space (i. e., a vector space along its inner product) to address both unsupervised and semi-supervised domain adaptation problems.
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.
Saliency Integration: An Arbitrator Model
Previous offline integration methods usually face two challenges: 1. if most of the candidate saliency models misjudge the saliency on an image, the integration result will lean heavily on those inferior candidate models; 2. an unawareness of the ground truth saliency labels brings difficulty in estimating the expertise of each candidate model.
Learning deep structured network for weakly supervised change detection
We apply a constrained mean-field algorithm to estimate the pixel-level labels, and use the estimated labels to update the parameters of the CNN in an iterative EM framework.
Going Deeper into Action Recognition: A Survey
Understanding human actions in visual data is tied to advances in complementary research areas including object recognition, human dynamics, domain adaptation and semantic segmentation.
Beyond Local Search: Tracking Objects Everywhere with Instance-Specific Proposals
Our method generates a small number of "high-quality" proposals by a novel instance-specific objectness measure and evaluates them against the object model that can be adopted from an existing tracking-by-detection approach as a core tracker.
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.
Pushing the Limits of Deep CNNs for Pedestrian Detection
In this work, we show that by re-using the convolutional feature maps (CFMs) of a deep convolutional neural network (DCNN) model as image features to train an ensemble of boosted decision models, we are able to achieve the best reported accuracy without using specially designed learning algorithms.
Class-Specific Image Deblurring
In image deblurring, a fundamental problem is that the blur kernel suppresses a number of spatial frequencies that are difficult to recover reliably.
Linearization to Nonlinear Learning for Visual Tracking
Building on the theory of globally linear approximations to nonlinear functions, we introduce an elegant method that jointly learns a nonlinear classifier and a visual dictionary for tracking objects in a semi-supervised sparse coding fashion.
Compositional Memory for Visual Question Answering
We then feed the episodes to a standard question answering module together with the contextual visual information and linguistic information.
Fast detection of multiple objects in traffic scenes with a common detection framework
The proposed framework consists of a dense feature extractor and detectors of three important classes.
When VLAD met Hilbert
Vectors of Locally Aggregated Descriptors (VLAD) have emerged as powerful image/video representations that compete with or even outperform state-of-the-art approaches on many challenging visual recognition tasks.
Tracking Randomly Moving Objects on Edge Box Proposals
Our method generates a small number of "high-quality" proposals by a novel instance-specific objectness measure and evaluates them against the object model that can be adopted from an existing tracking-by-detection approach as a core tracker.
Saliency-Aware Geodesic Video Object Segmentation
Building on the observation that foreground areas are surrounded by the regions with high spatiotemporal edge values, geodesic distance provides an initial estimation for foreground and background.
Ranked #5 on
Video Salient Object Detection
on DAVSOD-Difficult20
(using extra training data)
More About VLAD: A Leap From Euclidean to Riemannian Manifolds
This paper takes a step forward in image and video coding by extending the well-known Vector of Locally Aggregated Descriptors (VLAD) onto an extensive space of curved Riemannian manifolds.
A Novel Performance Evaluation Methodology for Single-Target Trackers
This paper addresses the problem of single-target tracker performance evaluation.
Efficient Upsampling of Natural Images
1) For the edge layer, we use a nonparametric approach by constructing a dictionary of patches from a given image, and synthesize edge regions in a higher-resolution version of the image.
DeepTrack: Learning Discriminative Feature Representations Online for Robust Visual Tracking
In this work, we present an efficient and very robust tracking algorithm using a single Convolutional Neural Network (CNN) for learning effective feature representations of the target object, in a purely online manner.
Super-Resolving Noisy Images
We then show that by taking a convex combination of orientation and frequency selective bands of the noisy and the denoised HR images, we can obtain a desired HR image where (i) some of the textural signal lost in the denoising step is effectively recovered in the HR domain, and (ii) additional textures can be easily synthesized by appropriately constraining the parameters of the convex combination.
Robust Orthonormal Subspace Learning: Efficient Recovery of Corrupted Low-rank Matrices
Low-rank matrix recovery from a corrupted observation has many applications in computer vision.
Bregman Divergences for Infinite Dimensional Covariance Matrices
We introduce an approach to computing and comparing Covariance Descriptors (CovDs) in infinite-dimensional spaces.
