Search Results for author:
Found 110 papers, 37 papers with code
Read Extensively, Focus Smartly: A Cross-document Semantic Enhancement Method for Visual Documents NER
To deal with this problem, this work proposes a cross-document semantic enhancement method, which consists of two modules: 1) To prevent distractions from irrelevant regions in the current document, we design a learnable attention mask mechanism, which is used to adaptively filter redundant information in the current document.
Single Domain Dynamic Generalization for Iris Presentation Attack Detection
To tackle the above issues, we propose a Single Domain Dynamic Generalization (SDDG) framework, which simultaneously exploits domain-invariant and domain-specific features on a per-sample basis and learns to generalize to various unseen domains with numerous natural images.
Taxonomy Completion with Probabilistic Scorer via Box Embedding
Specifically, TaxBox consists of three components: (1) a graph aggregation module to leverage the structural information of the taxonomy and two lightweight decoders that map features to box embedding and capture complex relationships between concepts; (2) two probabilistic scorers that correspond to attachment and insertion operations and ensure the avoidance of pseudo-leaves; and (3) three learning objectives that assist the model in mapping concepts more granularly onto the box embedding space.
Multi-view Adversarial Discriminator: Mine the Non-causal Factors for Object Detection in Unseen Domains
In this work, we present an idea to remove non-causal factors from common features by multi-view adversarial training on source domains, because we observe that such insignificant non-causal factors may still be significant in other latent spaces (views) due to the multi-mode structure of data.
Rethinking the Approximation Error in 3D Surface Fitting for Point Cloud Normal Estimation
Most existing approaches for point cloud normal estimation aim to locally fit a geometric surface and calculate the normal from the fitted surface.
1st Place Solution for ECCV 2022 OOD-CV Challenge Object Detection Track
The domain adaptation part is implemented as a Source-Free Domain Adaptation paradigm, which only uses the pre-trained model and the unlabeled target data to further optimize in a self-supervised training manner.
1st Place Solution for ECCV 2022 OOD-CV Challenge Image Classification Track
In the test-time training stage, we use the pre-trained model to assign noisy label for the unlabeled target data, and propose a Label-Periodically-Updated DivideMix method for noisy label learning.
Bit-Shrinking: Limiting Instantaneous Sharpness for Improving Post-Training Quantization
To this end, we detach a sharpness term from the loss which reflects the impact of quantization noise.
PHA: Patch-Wise High-Frequency Augmentation for Transformer-Based Person Re-Identification
Although recent studies empirically show that injecting Convolutional Neural Networks (CNNs) into Vision Transformers (ViTs) can improve the performance of person re-identification, the rationale behind it remains elusive.
NeRF-Gaze: A Head-Eye Redirection Parametric Model for Gaze Estimation
Gaze estimation is the fundamental basis for many visual tasks.
SAViT: Structure-Aware Vision Transformer Pruning via Collaborative Optimization
In this paper, we introduce joint importance, which integrates essential structural-aware interactions between components for the first time, to perform collaborative pruning.
Towards Comprehensive Representation Enhancement in Semantics-guided Self-supervised Monocular Depth Estimation
Semantics-guided self-supervised monocular depth estimation has been widely researched, owing to the strong cross-task correlation of depth and semantics.
Attention Diversification for Domain Generalization
Under this guidance, a novel Attention Diversification framework is proposed, in which Intra-Model and Inter-Model Attention Diversification Regularization are collaborated to reassign appropriate attention to diverse task-related features.
Multi-Scale Wavelet Transformer for Face Forgery Detection
To overcome these limitations, we propose a multi-scale wavelet transformer framework for face forgery detection.
Point Cloud Upsampling via Cascaded Refinement Network
In this manner, the proposed cascaded refinement network can be easily optimized without extra learning strategies.
FBNet: Feedback Network for Point Cloud Completion
The rapid development of point cloud learning has driven point cloud completion into a new era.
Unified Normalization for Accelerating and Stabilizing Transformers
To tackle these issues, we propose Unified Normalization (UN), which can speed up the inference by being fused with other linear operations and achieve comparable performance on par with LN.
TRIE++: Towards End-to-End Information Extraction from Visually Rich Documents
Most existing methods divide this task into two subparts: the text reading part for obtaining the plain text from the original document images and the information extraction part for extracting key contents.
E2-AEN: End-to-End Incremental Learning with Adaptively Expandable Network
In this paper, we propose an end-to-end trainable adaptively expandable network named E2-AEN, which dynamically generates lightweight structures for new tasks without any accuracy drop in previous tasks.
Dynamic Low-Resolution Distillation for Cost-Efficient End-to-End Text Spotting
In this paper, to address this problem, we propose a novel cost-efficient Dynamic Low-resolution Distillation (DLD) text spotting framework, which aims to infer images in different small but recognizable resolutions and achieve a better balance between accuracy and efficiency.
Knowledge Distillation
Optical Character Recognition (OCR)
+1
Semi-supervised Ranking for Object Image Blur Assessment
Based on this dataset, we propose a method to obtain the blur scores only with the pairwise rank labels as supervision.
Universal Domain Adaptive Object Detector
Universal domain adaptive object detection (UniDAOD)is more challenging than domain adaptive object detection (DAOD) since the label space of the source domain may not be the same as that of the target and the scale of objects in the universal scenarios can vary dramatically (i. e, category shift and scale shift).
Slimmable Domain Adaptation
In this paper, we introduce a simple framework, Slimmable Domain Adaptation, to improve cross-domain generalization with a weight-sharing model bank, from which models of different capacities can be sampled to accommodate different accuracy-efficiency trade-offs.
Label Matching Semi-Supervised Object Detection
To remedy this issue, we present a novel label assignment mechanism for self-training framework, namely proposal self-assignment, which injects the proposals from student into teacher and generates accurate pseudo labels to match each proposal in the student model accordingly.
2nd Place Solution for ICCV 2021 VIPriors Image Classification Challenge: An Attract-and-Repulse Learning Approach
Therefore, it is crucial to study how to learn more discriminative representations while avoiding over-fitting.
Transductive CLIP with Class-Conditional Contrastive Learning
This framework can reduce the impact of noisy labels from CLIP model effectively by combining both techniques.
Learning Domain Adaptive Object Detection with Probabilistic Teacher
In addition, we conduct anchor adaptation in parallel with localization adaptation, since anchor can be regarded as a learnable parameter.
Self-distilled Knowledge Delegator for Exemplar-free Class Incremental Learning
The knowledge extracted by the delegator is then utilized to maintain the performance of the model on old tasks in incremental learning.
KRNet: Towards Efficient Knowledge Replay
However, the number of stored latent codes in autoencoder increases linearly with the scale of data and the trained encoder is redundant for the replaying stage.
High-Efficiency Lossy Image Coding Through Adaptive Neighborhood Information Aggregation
To this end, Integrated Convolution and Self-Attention (ICSA) unit is first proposed to form a content-adaptive transform to characterize and embed neighborhood information dynamically of any input.
Few-shot One-class Domain Adaptation Based on Frequency for Iris Presentation Attack Detection
We thus define a new domain adaptation setting called Few-shot One-class Domain Adaptation (FODA), where adaptation only relies on a limited number of target bonafide samples.
Unimodal-Concentrated Loss: Fully Adaptive Label Distribution Learning for Ordinal Regression
In this paper, we emphatically summarize that learning an adaptive label distribution on ordinal regression tasks should follow three principles.
Weakly Supervised Regional and Temporal Learning for Facial Action Unit Recognition
Automatic facial action unit (AU) recognition is a challenging task due to the scarcity of manual annotations.
Few-Shot Class-Incremental Learning by Sampling Multi-Phase Tasks
In this work, we propose a new paradigm for FSCIL based on meta-learning by LearnIng Multi-phase Incremental Tasks (LIMIT), which synthesizes fake FSCIL tasks from the base dataset.
Ranked #4 on
Few-Shot Class-Incremental Learning
on CIFAR-100
class-incremental learning
Few-Shot Class-Incremental Learning
+2
End-to-End Modeling via Information Tree for One-Shot Natural Language Spatial Video Grounding
To achieve effective grounding under a limited annotation budget, we investigate one-shot video grounding, and learn to ground natural language in all video frames with solely one frame labeled, in an end-to-end manner.
Forward Compatible Few-Shot Class-Incremental Learning
Forward compatibility requires future new classes to be easily incorporated into the current model based on the current stage data, and we seek to realize it by reserving embedding space for future new classes.
Ranked #3 on
Few-Shot Class-Incremental Learning
on CIFAR-100
class-incremental learning
Few-Shot Class-Incremental Learning
+1
CAKE: A Scalable Commonsense-Aware Framework For Multi-View Knowledge Graph Completion
The previous knowledge graph embedding (KGE) techniques suffer from invalid negative sampling and the uncertainty of fact-view link prediction, limiting KGC's performance.
Learning Multiple Explainable and Generalizable Cues for Face Anti-spoofing
However, many other generalizable cues are unexplored for face anti-spoofing, which limits their performance under cross-dataset testing.
UWC: Unit-wise Calibration Towards Rapid Network Compression
This paper introduces a post-training quantization~(PTQ) method achieving highly efficient Convolutional Neural Network~ (CNN) quantization with high performance.
Cross-Modal ASR Post-Processing System for Error Correction and Utterance Rejection
Although modern automatic speech recognition (ASR) systems can achieve high performance, they may produce errors that weaken readers' experience and do harm to downstream tasks.
Automatic Speech Recognition
Automatic Speech Recognition (ASR)
+2
FAM: Visual Explanations for the Feature Representations From Deep Convolutional Networks
To handle this task, we propose a novel visual explanation paradigm called Feature Activation Mapping (FAM) in this paper.
Perform Like an Engine: A Closed-Loop Neural-Symbolic Learning Framework for Knowledge Graph Inference
Knowledge graph (KG) inference aims to address the natural incompleteness of KGs, including rule learning-based and KG embedding (KGE) models.
STEP: Out-of-Distribution Detection in the Presence of Limited In-Distribution Labeled Data
However, in many real-world applications, it is desirable to have SSL algorithms that not only classify the samples drawn from the same distribution of labeled data but also detect out-of-distribution (OOD) samples drawn from an unknown distribution.
Out-of-Distribution Detection
Out of Distribution (OOD) Detection
A Strong Baseline for Semi-Supervised Incremental Few-Shot Learning
Few-shot learning (FSL) aims to learn models that generalize to novel classes with limited training samples.
C+1 Loss: Learn to Classify C Classes of Interest and the Background Class Differentially
We figure out that the background class should be treated differently from the classes of interest during training.
Hindsight Reward Tweaking via Conditional Deep Reinforcement Learning
Designing optimal reward functions has been desired but extremely difficult in reinforcement learning (RL).
TransForensics: Image Forgery Localization with Dense Self-Attention
Nowadays advanced image editing tools and technical skills produce tampered images more realistically, which can easily evade image forensic systems and make authenticity verification of images more difficult.
Self-Supervised Regional and Temporal Auxiliary Tasks for Facial Action Unit Recognition
Based on these two self-supervised auxiliary tasks, local features, mutual relation and motion cues of AUs are better captured in the backbone network with the proposed regional and temporal based auxiliary task learning (RTATL) framework.
Divide-and-Assemble: Learning Block-wise Memory for Unsupervised Anomaly Detection
Surprisingly, by varying the granularity of division on feature maps, we are able to modulate the reconstruction capability of the model for both normal and abnormal samples.
InsPose: Instance-Aware Networks for Single-Stage Multi-Person Pose Estimation
Multi-person pose estimation is an attractive and challenging task.
Ranked #4 on
Multi-Person Pose Estimation
on COCO minival
RangeIoUDet: Range Image Based Real-Time 3D Object Detector Optimized by Intersection Over Union
Benefiting from the dense representation of the range image, RangeIoUDet is entirely constructed based on 2D convolution, making it possible to have a fast inference speed.
Entity Concept-enhanced Few-shot Relation Extraction
Few-shot relation extraction (FSRE) is of great importance in long-tail distribution problem, especially in special domain with low-resource data.
VSR: A Unified Framework for Document Layout Analysis combining Vision, Semantics and Relations
To address the above limitations, we propose a unified framework VSR for document layout analysis, combining vision, semantics and relations.
LGPMA: Complicated Table Structure Recognition with Local and Global Pyramid Mask Alignment
In this paper, we aim to obtain more reliable aligned bounding boxes by fully utilizing the visual information from both text regions in proposed local features and cell relations in global features.
Ranked #6 on
Table Recognition
on PubTabNet
Reciprocal Feature Learning via Explicit and Implicit Tasks in Scene Text Recognition
In this work, we excavate the implicit task, character counting within the traditional text recognition, without additional labor annotation cost.
RPVNet: A Deep and Efficient Range-Point-Voxel Fusion Network for LiDAR Point Cloud Segmentation
The voxel-based view is regular, but sparse, and computation grows cubically when voxel resolution increases.
Ranked #15 on
Robust 3D Semantic Segmentation
on SemanticKITTI-C
Modulating Localization and Classification for Harmonized Object Detection
Object detection involves two sub-tasks, i. e. localizing objects in an image and classifying them into various categories.
Self-Domain Adaptation for Face Anti-Spoofing
In this paper, we propose a self-domain adaptation framework to leverage the unlabeled test domain data at inference.
Multi-Level Adaptive Region of Interest and Graph Learning for Facial Action Unit Recognition
In order to incorporate the intra-level AU relation and inter-level AU regional relevance simultaneously, a multi-level AU relation graph is constructed and graph convolution is performed to further enhance AU regional features of each level.
Self-Supervised Noisy Label Learning for Source-Free Unsupervised Domain Adaptation
Usually, the given source domain pre-trained model is expected to optimize with only unlabeled target data, which is termed as source-free unsupervised domain adaptation.
Box Re-Ranking: Unsupervised False Positive Suppression for Domain Adaptive Pedestrian Detection
Therefore, it yields our attention to suppress false positive in each target domain in an unsupervised way.
SGMNet: Learning Rotation-Invariant Point Cloud Representations via Sorted Gram Matrix
Recently, various works that attempted to introduce rotation invariance to point cloud analysis have devised point-pair features, such as angles and distances.
Rethinking Pseudo-labeled Sample Mining for Semi-Supervised Object Detection
However, the impact of the pseudo-labeled samples' quality as well as the mining strategies for high quality training sample have rarely been studied in SSL.
A Free Lunch for Unsupervised Domain Adaptive Object Detection without Source Data
However, it is difficult to evaluate the quality of pseudo labels since no labels are available in target domain.
MANGO: A Mask Attention Guided One-Stage Scene Text Spotter
Recently end-to-end scene text spotting has become a popular research topic due to its advantages of global optimization and high maintainability in real applications.
Learning Open Set Network with Discriminative Reciprocal Points
In this process, one of the key challenges is to reduce the risk of generalizing the inherent characteristics of numerous unknown samples learned from a small amount of known data.
PolarDet: A Fast, More Precise Detector for Rotated Target in Aerial Images
Fast and precise object detection for high-resolution aerial images has been a challenging task over the years.
Ranked #32 on
Object Detection In Aerial Images
on DOTA
Joint Semantics and Data-Driven Path Representation for Knowledge Graph Inference
Inference on a large-scale knowledge graph (KG) is of great importance for KG applications like question answering.
AutoETER: Automated Entity Type Representation for Knowledge Graph Embedding
Recent advances in Knowledge Graph Embedding (KGE) allow for representing entities and relations in continuous vector spaces.
MAFF-Net: Filter False Positive for 3D Vehicle Detection with Multi-modal Adaptive Feature Fusion
Experimental results on the KITTI dataset demonstrate significant improvement in filtering false positive over the approach using only point cloud data.
RangeRCNN: Towards Fast and Accurate 3D Object Detection with Range Image Representation
We present RangeRCNN, a novel and effective 3D object detection framework based on the range image representation.
Two Step Joint Model for Drug Drug Interaction Extraction
When patients need to take medicine, particularly taking more than one kind of drug simultaneously, they should be alarmed that there possibly exists drug-drug interaction.
Drug–drug Interaction Extraction
named-entity-recognition
+4
Topic Adaptation and Prototype Encoding for Few-Shot Visual Storytelling
In this paper, we focus on enhancing the generalization ability of the VIST model by considering the few-shot setting.
Dynamic GCN: Context-enriched Topology Learning for Skeleton-based Action Recognition
The key lies in the design of the graph structure, which encodes skeleton topology information.
Learning a Domain Classifier Bank for Unsupervised Adaptive Object Detection
To tackle this issue, we develop a fine-grained domain alignment approach with a well-designed domain classifier bank that achieves the instance-level alignment respecting to their categories.
Text Recognition in Real Scenarios with a Few Labeled Samples
Scene text recognition (STR) is still a hot research topic in computer vision field due to its various applications.
Unsupervised Image Classification for Deep Representation Learning
Extensive experiments on ImageNet dataset have been conducted to prove the effectiveness of our method.
SPIN: Structure-Preserving Inner Offset Network for Scene Text Recognition
Arbitrary text appearance poses a great challenge in scene text recognition tasks.
Object-QA: Towards High Reliable Object Quality Assessment
In this paper, we define the problem of object quality assessment for the first time and propose an effective approach named Object-QA to assess high-reliable quality scores for object images.
TRIE: End-to-End Text Reading and Information Extraction for Document Understanding
Since real-world ubiquitous documents (e. g., invoices, tickets, resumes and leaflets) contain rich information, automatic document image understanding has become a hot topic.
IROS 2019 Lifelong Robotic Vision Challenge -- Lifelong Object Recognition Report
This report summarizes IROS 2019-Lifelong Robotic Vision Competition (Lifelong Object Recognition Challenge) with methods and results from the top $8$ finalists (out of over~$150$ teams).
Counterfactual Samples Synthesizing for Robust Visual Question Answering
To reduce the language biases, several recent works introduce an auxiliary question-only model to regularize the training of targeted VQA model, and achieve dominating performance on VQA-CP.
Ranked #1 on
Visual Question Answering (VQA)
on VQA-CP
(using extra training data)
Neural Inheritance Relation Guided One-Shot Layer Assignment Search
In this paper, for the first time, we systematically investigate the impact of different layer assignments to the network performance by building an architecture dataset of layer assignment on CIFAR-100.
Refined Gate: A Simple and Effective Gating Mechanism for Recurrent Units
Recurrent neural network (RNN) has been widely studied in sequence learning tasks, while the mainstream models (e. g., LSTM and GRU) rely on the gating mechanism (in control of how information flows between hidden states).
Text Perceptron: Towards End-to-End Arbitrary-Shaped Text Spotting
Many approaches have recently been proposed to detect irregular scene text and achieved promising results.
An End-to-End Audio Classification System based on Raw Waveforms and Mix-Training Strategy
This paper introduces an end-to-end audio classification system based on raw waveforms and mix-training strategy.
Fast Task Adaptation for Few-Shot Learning
The key lies in generalization of prior knowledge learned from large-scale base classes and fast adaptation of the classifier to novel classes.
Adversarial Seeded Sequence Growing for Weakly-Supervised Temporal Action Localization
The second module is a specific classifier for mining trivial or incomplete action regions, which is trained on the shared features after erasing the seeded regions activated by SSG.
Action Detection
Weakly-supervised Temporal Action Localization
+1
Learned Quality Enhancement via Multi-Frame Priors for HEVC Compliant Low-Delay Applications
Networked video applications, e. g., video conferencing, often suffer from poor visual quality due to unexpected network fluctuation and limited bandwidth.
Posterior-regularized REINFORCE for Instance Selection in Distant Supervision
This paper provides a new way to improve the efficiency of the REINFORCE training process.
All You Need is a Few Shifts: Designing Efficient Convolutional Neural Networks for Image Classification
In this family of architectures, the basic block is only composed by 1x1 convolutional layers with only a few shift operations applied to the intermediate feature maps.
You Only Recognize Once: Towards Fast Video Text Spotting
Video text spotting is still an important research topic due to its various real-applications.
Collaborative Spatio-temporal Feature Learning for Video Action Recognition
By sharing the convolution kernels of different views, spatial and temporal features are collaboratively learned and thus benefit from each other.
Action Recognition In Videos
Temporal Action Localization
+1
Cross-relation Cross-bag Attention for Distantly-supervised Relation Extraction
Distant supervision leverages knowledge bases to automatically label instances, thus allowing us to train relation extractor without human annotations.
Learning Incremental Triplet Margin for Person Re-identification
In particular, we propose a novel multi-stage training strategy which learns incremental triplet margin and improves triplet loss effectively.
A Layer Decomposition-Recomposition Framework for Neuron Pruning towards Accurate Lightweight Networks
A better alternative is to propagate the entire useful information to reconstruct the pruned layer instead of directly discarding the less important neurons.
Counterfactual Critic Multi-Agent Training for Scene Graph Generation
CMAT is a multi-agent policy gradient method that frames objects as cooperative agents, and then directly maximizes a graph-level metric as the reward.
Segregated Temporal Assembly Recurrent Networks for Weakly Supervised Multiple Action Detection
Finally, we transform the output of recurrent neural network into the corresponding action distribution.
Small-scale Pedestrian Detection Based on Topological Line Localization and Temporal Feature Aggregation
A critical issue in pedestrian detection is to detect small-scale objects that will introduce feeble contrast and motion blur in images and videos, which in our opinion should partially resort to deep-rooted annotation bias.
Extreme Network Compression via Filter Group Approximation
In this paper we propose a novel decomposition method based on filter group approximation, which can significantly reduce the redundancy of deep convolutional neural networks (CNNs) while maintaining the majority of feature representation.
Small-scale Pedestrian Detection Based on Somatic Topology Localization and Temporal Feature Aggregation
A critical issue in pedestrian detection is to detect small-scale objects that will introduce feeble contrast and motion blur in images and videos, which in our opinion should partially resort to deep-rooted annotation bias.
Ranked #17 on
Pedestrian Detection
on CityPersons
A practical convolutional neural network as loop filter for intra frame
It aims to design a single CNN model with low redundancy to adapt to decoded frames with different qualities and ensure consistency.
Multimedia
Edit Probability for Scene Text Recognition
The advantage lies in that the training process can focus on the missing, superfluous and unrecognized characters, and thus the impact of the misalignment problem can be alleviated or even overcome.
Co-occurrence Feature Learning from Skeleton Data for Action Recognition and Detection with Hierarchical Aggregation
Skeleton-based human action recognition has recently drawn increasing attentions with the availability of large-scale skeleton datasets.
Ranked #2 on
Skeleton Based Action Recognition
on PKU-MMD
AON: Towards Arbitrarily-Oriented Text Recognition
Existing methods on text recognition mainly work with regular (horizontal and frontal) texts and cannot be trivially generalized to handle irregular texts.
Ranked #8 on
Scene Text Recognition
on ICDAR 2003
Cascade Region Proposal and Global Context for Deep Object Detection
Deep region-based object detector consists of a region proposal step and a deep object recognition step.
Focusing Attention: Towards Accurate Text Recognition in Natural Images
FAN consists of two major components: an attention network (AN) that is responsible for recognizing character targets as in the existing methods, and a focusing network (FN) that is responsible for adjusting attention by evaluating whether AN pays attention properly on the target areas in the images.
Skeleton-based Action Recognition with Convolutional Neural Networks
Current state-of-the-art approaches to skeleton-based action recognition are mostly based on recurrent neural networks (RNN).
Ranked #3 on
Skeleton Based Action Recognition
on PKU-MMD
All You Need is Beyond a Good Init: Exploring Better Solution for Training Extremely Deep Convolutional Neural Networks with Orthonormality and Modulation
Moreover, we can successfully train plain CNNs to match the performance of the residual counterparts.
Mixed context networks for semantic segmentation
Semantic segmentation is challenging as it requires both object-level information and pixel-level accuracy.
