Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileAcademic lineage
Possible advisorsa guess from early papers, not confirmed
- Qi TianSuggested from co-authorship
- Longin Jan LateckiSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works140 from public data
- CenterNet: Keypoint Triplets for Object Detection3,591
This paper presents an efficient solution that explores the visual patterns within individual cropped regions with minimal costs, and builds the framework upon a representative one-stage keypoint-based detector named CornerNet, which improves both precision and recall.
- CenterNet: Object Detection with Keypoint Triplets3,563
This paper presents an efficient solution that explores the visual patterns within individual cropped regions with minimal costs, and builds the framework upon a representative one-stage keypoint-based detector named CornerNet, which improves both precision and recall.
- Improving Transferability of Adversarial Examples With Input Diversity1,524
This work proposes to improve the transferability of adversarial examples by creating diverse input patterns by applying random transformations to the input images at each iteration, and shows that the proposed attack method can generate adversarialExamples that transfer much better to different networks than existing baselines.
- Hypergraph convolution and hypergraph attention952
This work introduces two end-to-end trainable operators to the family of graph neural networks, i.e., hypergraph convolution and hypergraph attention, which define the basic formulation of performing convolution on a hypergraph and further enhances the capacity of representation learning by leveraging an attention module.
- Asymmetric Non-Local Neural Networks for Semantic Segmentation712
Asymmetric Pyramid Non- local Block (APNB) and Asymmetric Fusion Non-local Block (AFNB) are presented, which has two prominent components: APNB leverages a pyramid sampling module into the non-local block to largely reduce the computation and memory consumption without sacrificing the performance.
- DanceTrack: Multi-Object Tracking in Uniform Appearance and Diverse Motion467
A large-scale dataset for multi-human tracking, where humans have sim-ilar appearance, diverse motion and extreme articulation, and is named “DanceTrack”, which is expected to provide a better platform to develop more MOT algorithms that rely less on visual discrimination and depend more on motion analysis.
- Is synthetic data from generative models ready for image recognition?443
This work extensively study whether and how synthetic images generated from state-of-the-art text-to-image generation models can be used for image recognition tasks, and focuses on two perspectives: synthetic data for improving classification models in data-scarce settings (i.e. zero-shot and few-shot), and syntheticData for large-scale model pre-training for transfer learning.
- 440
- PCL: Proposal Cluster Learning for Weakly Supervised Object Detection422
This paper first shows that instances can be assigned object or background labels directly based on proposal clusters for instance classifier refinement, and then shows that treating each cluster as a small new bag yields fewer ambiguities than the directly assigning label method.
- DragDiffusion: Harnessing Diffusion Models for Interactive Point-Based Image Editing393
This work introduces two additional techniques, namely identity-preserving fine-tuning and reference-latent-control, to further preserve the identity of the original image and presents a challenging benchmark dataset called DRAGBENCH─ the first benchmark to evaluate the performance of interactive point-based image editing methods.
- Triplet-Center Loss for Multi-view 3D Object Retrieval385
This paper study variants of deep metric learning losses for 3D object retrieval, which did not receive enough attention from this area, and proposes a novel loss named triplet-center loss, which can further enhance the discriminative power of the features.
- 373
- Scalable Person Re-identification on Supervised Smoothed Manifold342
An unconventional manifold-preserving algorithm is proposed, which can make best use of supervision from training data, whose label information is given as pairwise constraints, scale up to large repositories with low on-line time complexity, and be plunged into most existing algorithms, serving as a generic postprocessing procedure to further boost the identification accuracies.
- End-to-End Temporal Action Detection With Transformer341
TadTR is an end-to-end Transformer-based method for temporal action detection that achieves state-of-the-art performance on THUMOS14 and HACS Segments, and requires lower computation cost than previous detectors, while preserving remarkable performance.
- MOSE: A New Dataset for Video Object Segmentation in Complex Scenes307
The experiments show that current VOS algorithms cannot well perceive objects in complex scenes, and there are unresolved challenges under complex scenes and more efforts are desired to explore these challenges in the future.
- GIFT: A Real-Time and Scalable 3D Shape Search Engine298
The proposed 3D shape search engine, which combines GPU acceleration and Inverted File (Twice), is named as GIFT, which outperforms the state-of-the-art methods significantly in retrieval accuracy on various shape benchmarks and competitions.
- PLA: Language-Driven Open-Vocabulary 3D Scene Understanding240
This work proposes to distill knowledge encoded in pretrained vision-language (VL) foundation models through captioning multi-view images from 3D, which allows explicitly associating 3D and semantic-rich captions in open-vocabulary captions.
- Occluded Video Instance Segmentation: A Benchmark229
A simple plug-and-play module that performs temporal feature calibration to complement missing object cues caused by occlusion is presented, and a remarkable AP improvement on the OVIS dataset is obtained.
- Prior-Aware Neural Network for Partially-Supervised Multi-Organ Segmentation204
PaNN achieves state-of-the-art performance on the MICCAI2015 challenge ``Multi-Atlas Labeling Beyond the Cranial Vault'', a competition on organ segmentation in the abdomen, and is reformulated as a min-max form and optimized via the stochastic primal-dual gradient algorithm.
- XingGAN for Person Image Generation197
A novel Generative Adversarial Network (XingGAN or CrossingGAN) for person image generation tasks, i.e., translating the pose of a given person to a desired one to mutually improve each other is proposed.
- SwiftNet: Real-time Video Object Segmentation190
This work presents SwiftNet for real-time semisupervised video object segmentation (one-shot VOS), which reports 77.8% $\mathcal{J}\& \mathcal {F}$ and 70 FPS on DAVIS 2017 validation dataset, leading all present solutions in overall accuracy and speed performance.
- Deep Learning Representation using Autoencoder for 3D Shape Retrieval187
This work projects 3D shapes into 2D space and uses autoencoder for feature learning on the 2D images and shows the proposed deep learning feature is complementary to conventional local image descriptors, which can obtain the state-of-the-art performance on 3D shape retrieval benchmarks.
- Learning Transferable Adversarial Examples via Ghost Networks171
Ghost Networks is proposed to improve the transferability of adversarial examples by reproducing the NeurIPS 2017 adversarial competition, and outperforms the No.1 attack submission by a large margin, demonstrating its effectiveness and efficiency.
- Learning Attraction Field Representation for Robust Line Segment Detection169
A region-partition based attraction field dual representation for line segment maps, which poses the problem of line segment detection (LSD) as the region coloring problem and harnesses the best practices developed in ConvNets based semantic segmentation methods such as the encoder-decoder architecture and the a-trous convolution.
- Symmetry-Constrained Rectification Network for Scene Text Recognition166
A Symmetry-constrained Rectification Network (ScRN) based on local attributes of text instances, such as center line, scale and orientation, which enables ScRN to generate better rectification results than existing methods and thus lead to higher recognition accuracy.
- Semi-Supervised 3D Abdominal Multi-Organ Segmentation Via Deep Multi-Planar Co-Training166
Deep Multi-Planar Co-Training is proposed, whose contributions can be divided into two folds: 1) the deep model is learned in a co-training style which can mine consensus information from multiple planes like the sagittal, coronal, and axial planes; 2) Multi-planar fusion is applied to generate more reliable pseudo-labels, which alleviates the errors occurring in the pseudo-Labels and thus can help to train better segmentation networks.
- SeqFormer: Sequential Transformer for Video Instance Segmentation163
This work presents SeqFormer, a model that locates an instance in each frame and aggregates temporal information to learn a powerful representation of a video-level instance, which is used to predict the mask sequences on each frame dynamically.
- 161
- Learn to Scale: Generating Multipolar Normalized Density Maps for Crowd Counting157
First, a patch-level density map is extracted by a density estimation model and further grouped into several density levels which are determined over full datasets, which can significantly condense the density distribution into several clusters, and enable that the density variance can be learned by a single model.
- Hard-Aware Point-to-Set Deep Metric for Person Re-identification156
In addition to person re-ID, the proposed Hard-Aware Point-to-Set (HAP2S) loss with a soft hard-mining scheme applies to generic deep metric learning benchmarks including CUB-200-2011 and Cars196, and achieves state-of-the-art results.
- Holistically-Attracted Wireframe Parsing143
This paper presents a fast and parsimonious parsing method to accurately and robustly detect a vectorized wireframe in an input image with a single forward pass, and is thus called Holistically-Attracted Wireframe Parser (HAWP).
- 142
- TransMix: Attend to Mix for Vision Transformers140
TransMix is proposed, which mixes labels based on the attention maps of Vision Transformers, which can consistently improve various ViT-based models at scales on ImageNet classification.
- AutoScale: Learning to Scale for Crowd Counting137
A simple and effective Learning to Scale (L2S) module, which automatically scales dense regions into reasonable closeness levels (reflecting image-plane distance between neighboring people), and introduces a customized dynamic cross-entropy loss, significantly improving the localization-based model optimization.
- Anchor-Free Person Search133
This work presents the Feature-Aligned Person Search Network (AlignPS), the first anchor-free framework to efficiently tackle this challenging task, and proposes an aligned feature aggregation module to generate more discriminative and robust feature embeddings by following a "re-id first" principle.
- Anchor Diffusion for Unsupervised Video Object Segmentation128
Inspired by the non-local operators, a technique to establish dense correspondences between pixel embeddings of a reference "anchor" frame and the current one is introduced, which allows the learning of pairwise dependencies at arbitrarily long distances without conditioning on intermediate frames.
- General Object Foundation Model for Images and Videos at Scale120
GLEE is presented, an object-level foundation model for locating and identifying objects in images and videos, capable of being integrated into Large Language Models, serving as a foun-dational model to provide universal object-level information for multimodal tasks.
- Corner Proposal Network for Anchor-Free, Two-Stage Object Detection118
A novel anchor-free, two-stage framework which first extracts a number of object proposals by finding potential corner keypoint combinations and then assigns a class label to each proposal by a standalone classification stage is proposed, which is competitive among state-of-the-art object detection methods.
- CenterNet++ for Object Detection112
It is demonstrated that bottom-up approaches show competitive performance compared with top-down approaches and have higher recall rates and a real-time CenterNet model, an anchor-free detector that achieves a good trade-off between accuracy and speed.
- 110
- 105
- Multi-shot Temporal Event Localization: a Benchmark102
This paper proposes a new and challenging task called multi-shot temporal event localization, and accordingly, collects a large-scale dataset called MUlti-Shot EventS (MUSES), which has 31,477 event instances for a total of 716 video hours.
- 98
- 98
- Divide and Fuse: A Re-ranking Approach for Person Re-identification96
A "Divide and use" re- ranking framework for person re-ID that exploits the diversity from different parts of a high-dimensional feature vector for fusion-based re-ranking, while no other features are accessible.
- 94
- Dual Attention GANs for Semantic Image Synthesis92
A novel Dual Attention GAN (DAGAN) is proposed to synthesize photo-realistic and semantically-consistent images with fine details from the input layouts without imposing extra training overhead or modifying the network architectures of existing methods.
- 92
- Towards Understanding and Mitigating Dimensional Collapse in Heterogeneous Federated Learning89
This work proposes {\sc FedDecorr], a novel method that can effectively mitigate dimensional collapse in federated learning that applies a regularization term during local training that encourages different dimensions of representations to be uncorrelated.
- Mimicking the Oracle: An Initial Phase Decorrelation Approach for Class Incremental Learning89
Class-wise Decorrelation (CwD) is proposed that regularizes representations of each class to scatter more uniformly, thus mimicking the model jointly trained with all classes (i.e., the oracle model), and is simple to implement and easy to plug into existing methods.
- Advancing COVID-19 diagnosis with privacy-preserving collaboration in artificial intelligence82
A privacy-preserving AI framework for CT-based COVID-19 diagnosis is presented and demonstrated on data from 23 hospitals in China and the United Kingdom, showing that the federated learning framework model considerably outperformed all of the local models.
- 81
- An Empirical Study of End-to-End Temporal Action Detection77
A mid-resolution baseline detector is built, which achieves the state-of-the-art performance of end-to-end temporal action detection methods while running more than 4×faster.
- 77
- MOSEv2: A More Challenging Dataset for Video Object Segmentation in Complex Scenes74
This work presents MOSEv2, a significantly more challenging dataset designed to further advance VOS methods under real-world conditions, and benchmark 20 representative VOS methods under 5 different settings, highlighting that current VOS methods still fall short under real-world complexities.
- 71
- Regularized Diffusion Process for Visual Retrieval70
A new variant o diffusion process, which also operates on a tensor product graph is proposed, defined in three equivalent formulations (regularization framework, iterative framework and limit framework, respectively), which theoretically explain how this kind of diffusion process can better reveal the intrinsic relationship between objects.
- View N-Gram Network for 3D Object Retrieval68
Inspired by n-gram models in natural language processing, VNN divides the view sequence into a set of visual n- grams, which involve overlapping consecutive view sub-sequences, which helps to learn a discriminative global embedding for each 3D object.
- Learning to Holistically Detect Bridges From Large-Size VHR Remote Sensing Imagery65
This paper proposes a large-scale dataset named GLH-Bridge comprising 6,000 VHR RSIs sampled from diverse geographic locations across the globe and presents an efficient network for holistic bridge detection (HBD-Net) in large-size RSIs.
- Importance-Aware Semantic Segmentation in Self-Driving with Discrete Wasserstein Training65
This paper proposes to incorporate the importance-aware inter-class correlation in a Wasserstein training framework by configuring its ground distance matrix by extending the ground metric to a linear, convex or concave increasing function w.r.t. pre-defined ground distance.
- Bipartite Graph Reasoning GANs for Person Image Generation62
A novel Bipartite Graph Reasoning GAN (BiGraphGAN) block aims to reason the crossing long-range relations between the source pose and the target pose in a bipartite graph, which mitigates some challenges caused by pose deformation.
- Large-Scale 3D Shape Retrieval from ShapeNet Core5561
This track aims to provide a benchmark to evaluate large-scale shape retrieval based on the ShapeNet dataset, using ShapeNet Core55, which provides more than 50 thousands models over 55 common categories in total for training and evaluating several algorithms.
- Regional Homogeneity: Towards Learning Transferable Universal Adversarial Perturbations Against Defenses59
It is shown that a simple universal perturbation can fool a series of state-of-the-art defenses, and it is verified that regionally homogeneous perturbations can well transfer across different vision tasks.
- 58
- 58
- Lowis3D: Language-Driven Open-World Instance-Level 3D Scene Understanding57
To tackle the localization challenge for novel classes in the open-world setting, this work develops debiased instance localization, which significantly improves the generalization capabilities of instance grouping and, thus, the ability to accurately locate novel objects.
- Neural Architecture Search for Lightweight Non-Local Networks55
This work proposes a Lightweight Non-Local (LightNL) block by squeezing the transformation operations and incorporating compact features, and proposes an efficient neural architecture search algorithm to learn an optimal configuration of LightNL blocks in an end-to-end manner.
- Knowledge Distillation as Efficient Pre-training: Faster Convergence, Higher Data-efficiency, and Better Transferability53
This work investigates an alternative strategy for pre-training, namely Knowledge Distillation as Efficient Pre-training (KDEP), aiming to efficiently transfer the learned feature representation from existing pre-trained models to new student models for future downstream tasks.
- PlaneTR: Structure-Guided Transformers for 3D Plane Recovery52
A neural network built upon Transformers, namely PlaneTR, to simultaneously detect and reconstruct planes from a single image and achieves state-of-the-art performance on the ScanNet and NYUv2 datasets.
- Language Matters: A Weakly Supervised Vision-Language Pre-training Approach for Scene Text Detection and Spotting51
A weakly supervised pre-training method, oCLIP, which can acquire effective scene text representations by jointly learning and aligning visual and textual information and outperforms existing pre- training techniques consistently across multiple public datasets.
- Fourier Document Restoration for Robust Document Dewarping and Recognition44
FDRNet is presented, a Fourier Document Restoration Network that can restore documents with different distortions and improve document recognition in a reliable and simpler manner and outperforms the state-of-the-art by large margins on both dewarping and text recognition tasks.
- Beyond diffusion process: Neighbor set similarity for fast re-ranking44
A simple yet effective method called Neighbor Set Similarity (NSS) is proposed, which is superior to both traditional pairwise similarity and diffusion process, and suggests that NSS is more suitable for large scale image retrieval than diffusion process.
- Learning Regional Attraction for Line Segment Detection39
The problem of line segment detection (LSD) as a problem of region coloring is posed, and an end-to-end framework to learn attraction field maps for raw input images, followed by a squeeze module to detect line segments is developed.
- 35
- Smooth Neighborhood Structure Mining on Multiple Affinity Graphs with Applications to Context-Sensitive Similarity34
A new algorithm called Smooth Neighborhood (SN) is proposed that mines the neighborhood structure to satisfy the manifold assumption and is adjusted to tackle multiple affinity graphs by imposing a weight learning paradigm.
- 28
- PV3D: A 3D Generative Model for Portrait Video Generation28
Pv3D is proposed, the first generative framework that can synthesize multi-view consistent portrait videos with high-quality appearance and geometry and is able to support many downstream applications such as animating static portraits and view-consistent video motion editing.
- 28
- 28
- 28
- Neural shape codes for 3D model retrieval28
This work proposes a practical and effective way for applying Convolutional Neural Network to 3D model retrieval, by training the network with the depth projections of3D model.
- Image-to-Character-to-Word Transformers for Accurate Scene Text Recognition27
The proposed I2C2W outperforms the state-of-the-art by large margins for challenging scene text dataset with various curvature and perspective distortions and achieves very competitive recognition performance over multiple normal scene text datasets.
- 27
- 26
- 25
- DIRECT-3D: Learning Direct Text-to-3D Generation on Massive Noisy 3D Data23
DIRECT-3D is a diffusion-based 3D generative model for creating high-quality 3D assets from text prompts that achieves state-of-the-art performance in both single-class generation and text-to-3D generation and can serve as a useful 3D geometric prior of objects.
- 22
- Multidimensional Scaling on Multiple Input Distance Matrices22
This paper proposes a new algorithm called Multi-View Multidimensional Scaling (MVMDS) by considering each input distance matrix as one view, which can learn the weights of views automatically by exploring the consensus information and complementary nature of views.
- Dataset Condensation via Generative Model21
This work proposes to condense the dataset into another format, a generative model, and is the first to successfully conduct condensation on ImageNet-1k.
- Timeexpert: an Expert-Guided Video Llm for Video Temporal Grounding19
TimeExpert is introduced, a Mixture-of-Experts (MoE)-based Video-LLM that effectively decomposes VTG tasks by dynamically routing task-specific tokens to specialized experts, with increased computational efficiency.
- Feature context learning for human parsing19
This paper proposes an end-to-end trainable, highly flexible and generic module called feature context module (FCM), which explores the correlation of adjacent pixels and aggregates the contextual information embedded in the real topology of the human body.
- Metric Attack and Defense for Person Re-identification17
This work proposes Adversarial Metric Attack, a parallel methodology to adversarial classification attacks, and presents an early attempt of training a metric-preserving network, thereby defending the metric against adversarial attacks.
- 17
- 17
- 16
- Discovering Failure Modes of Text-guided Diffusion Models via Adversarial Search14
This work proposes SAGE, an adversarial attack on TDMs that uses image classifiers as surrogate loss functions, to search over the discrete prompt space and the high-dimensional latent space of TDMs to automatically discover unexpected behaviors and failure cases in the image generation.
- 2D Scene Sketch-Based 3D Scene Retrieval14
The objective of this track is to evaluate the performance of different 2D scene sketch-based 3D scene retrieval algorithms using a 2D sketch query dataset and a 3D Warehouse model dataset and suggest several future research directions to tackle this research problem.
- Debiasing Text-to-Image Diffusion Models13
This work formalizes the problem setting and uses the text descriptions of bias groups to establish an unsafe direction for guiding the diffusion process, and proposes an iterative distribution alignment (IDA) method, which shows efficiency and fast convergence in resolving the social bias in TTI diffusion models.
- 12
- 12
- Improving context-sensitive similarity via smooth neighborhood for object retrieval12
A new algorithm called Smooth Neighborhood (SN) is proposed that mines the neighborhood structure to satisfy the manifold assumption and can yield state-of-the-art performances on shape retrieval, image retrieval and 3D model retrieval.
- Training convolutional neural network from multi-domain contour images for 3D shape retrieval12
This work takes 2D sketches and edge maps of 2D views of 3D models as a unified domain to train the Convolutional Neural Network (CNN), and shows that the CNN can successfully learn the object structure from different types of clues.
- Multiple Stage Residual Model for Accurate Image Classification12
This paper proposes an effective and efficient model, which is called Multiple Stage Residual Model (MSRM), to make full use of the residual vector to generate a multiple stage code.
- Text4Seg++: Advancing Image Segmentation via Generative Language Modeling11
A novel text-as-mask paradigm that casts image segmentation as a text generation problem, eliminating the need for additional decoders and significantly simplifying the segmentation process is proposed, highlighting the effectiveness, scalability, and generalizability of text-driven image segmentation within the MLLM framework.
- LATFormer: Locality-Aware Point-View Fusion Transformer for 3D shape recognition11
This work proposes a novel Locality-Aware Point-View Fusion Transformer (LATFormer) for 3D shape retrieval and classification and utilizes the LAF module to fuse the multi-scale features of the two modalities both bidirectionally and hierarchically to obtain more informative features.
- A comparison of methods for 3D scene shape retrieval11
A comprehensive comparison of all the above methods and an additional new retrieval methods are performed by evaluating them on the two benchmarks to solicit state-of-the-art approaches.
- Progress and Prospects in 3D Generative AI: A Technical Overview including 3D human10
The AI generated object models in 3D, followed by the generated 3D human models, and finally, the generated 3D human motions are discussed, culminating in a conclusive summary and a vision for the future.
- Free-ATM: Exploring Unsupervised Learning on Diffusion-Generated Images with Free Attention Masks9
This work uncovers that diffusion models' cross-attention layers inherently provide annotation-free attention masks aligned with corresponding text inputs on generated images that can close the performance gap between unsupervised pretraining on synthetic data and real-world scenarios.
- InstMove: Instance Motion for Object-centric Video Segmentation9
InstMove, which stands for Instance Motion for Object-centric Video Segmentation, mainly relies on instance-level motion information that is free from image feature embeddings, and features physical interpretations, making it more accurate and robust toward occlusion and fast-moving objects.
- LUMix: Improving Mixup by Better Modelling Label Uncertainty8
This paper proposes LUMix, which models such uncertainty by adding label perturbation during training and can be universally applied to any deep networks, with minimal computational cost.
- 8
- Adversarial Metric Attack for Person Re-identification.8
This work proposes Adversarial Metric Attack, a parallel methodology to adversarial classification attacks, which can effectively generate adversarial examples for re-ID, and expects that this work can facilitate the development of robust feature learning with the experimental conclusions.
- 7
- 7
- 7
- 6
- Describe, Adapt and Combine: Empowering CLIP Encoders for Open-Set 3D Object Retrieval6
A simple yet effective framework named Describe, Adapt and Combine (DAC) is presented, which innovatively synergizes a CLIP model with a multi-modal large language model (MLLM) to learn generalized 3D representations, where the MLLM is used for dual purposes.
- 6
- Deep Interactive Video Inpainting: An Invisibility Cloak for Harry Potter6
This is the first deep learning-based interactive video inpainting framework that only uses a free form of user input as guidance (i.e. scribbles) instead of mask annotations, which has academic, entertainment, and commercial value.
- 6
- 6
- 5
- Co-spectral for robust shape clustering5
A similarity fusion framework based on spectral analysis is proposed, which can automatically make use of the complementarity of various shape similarities based on a co-training framework and outperforms the state-of-the-art algorithms by a large margin.
- 4
- 4
- 4
- 2
- 1
- Runner-Up Solution to ECCV 2022 Challenge on Out of Vocabulary Scene Text Understanding: Cropped Word Recognition1
This report presents the 2nd place solution to ECCV 2022 challenge on Out-of-Vocabulary Scene Text Understanding (OOV-ST) : Cropped Word Recognition, which achieves a word accuracy of 59.45\% when considering out- of-vocabulary words only.
- 1
- 1
- 1
- MIS '24: 1st ACM Multimedia Workshop on Multi-modal Misinformation Governance in the Era of Foundation Models–
This workshop brings together researchers and practitioners to discuss key topics in multi-modal misinformation governance, including new datasets, evaluation techniques, ethical considerations, methodological progress, case studies, and future research directions.
- –
- –
- The Runner-up Solution for YouTube-VIS Long Video Challenge 2022–
This technical report describes the 2nd-place solution for the ECCV 2022 YouTube-VIS Long Video Challenge, which adopts the previously proposed online video instance segmentation method IDOL and uses pseudo labels to help contrastive learning.
- 1st Place Solution to ECCV 2022 Challenge on Out of Vocabulary Scene Text Understanding: End-to-End Recognition of Out of Vocabulary Words–
This report describes the solution to the Out of Vocabulary Scene Text Understanding (OOV-ST) Challenge, which aims to extract out-of-vocabulary words from natural scene images, and achieves 1st in end-to-end OOV word recognition track of OOV Challenge in ECCV2022 TiE Workshop.
- –
- –
- –
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-12. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.