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
- Shuicheng YanSuggested from co-authorship
Is this you? Claim this profile to confirm or dismiss it.
Works276 from public data
- CCNet: Criss-Cross Attention for Semantic Segmentation3,111
For each pixel, a novel criss-cross attention module in CCNet harvests the contextual information of all the pixels on its criss-cross path by taking a further recurrent operation, each pixel can finally capture the full-image dependencies from all pixels.
- 939
- Object Region Mining with Adversarial Erasing: A Simple Classification to Semantic Segmentation Approach900
This work investigates a principle way to progressively mine discriminative object regions using classification networks to address the weakly-supervised semantic segmentation problems and proposes a new adversarial erasing approach for localizing and expanding object regions progressively.
- Perceptual Generative Adversarial Networks for Small Object Detection888
This work addresses the small object detection problem by developing a single architecture that internally lifts representations of small objects to super-resolved ones, achieving similar characteristics as large objects and thus more discriminative for detection.
- HCP: A Flexible CNN Framework for Multi-Label Image Classification827
Experimental results on Pascal VOC 2007 and VOC 2012 multi-label image datasets well demonstrate the superiority of the proposed HCP infrastructure over other state-of-the-arts, where an arbitrary number of object segment hypotheses are taken as the inputs.
- Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi-Supervised Semantic Segmentation648
It is found that varying dilation rates can effectively enlarge the receptive fields of convolutional kernels and more importantly transfer the surrounding discriminative information to non-discriminative object regions, promoting the emergence of these regions in the object localization maps.
- Adversarial Complementary Learning for Weakly Supervised Object Localization626
This work mathematically proves that class localization maps can be obtained by directly selecting the class-specific feature maps of the last convolutional layer, which paves a simple way to identify object regions and presents a simple network architecture including two parallel-classifiers for object localization.
- STC: A Simple to Complex Framework for Weakly-Supervised Semantic Segmentation617
A simple to complex (STC) framework in which only image-level annotations are utilized to learn DCNNs for semantic segmentation, which demonstrates the superiority of the proposed STC framework compared with other state-of-the-arts frameworks.
- SG-One: Similarity Guidance Network for One-Shot Semantic Segmentation599
This article proposes a simple yet effective similarity guidance network to tackle the one-shot (SG-One) segmentation problem, aiming at predicting the segmentation mask of a query image with the reference to one densely labeled support image of the same category.
- 563
- Horizontal Pyramid Matching for Person Re-Identification512
A simple yet effective Horizontal Pyramid Matching (HPM) approach to fully exploit various partial information of a given person, so that correct person candidates can be still identified even even some key parts are missing.
- University-1652: A Multi-view Multi-source Benchmark for Drone-based Geo-localization505
It is argued that drones could serve as the third platform to deal with the geo-localization problem and propose a strong CNN baseline on this challenging dataset, named University-1652, which is the first drone-based geo- localization dataset and enables two new tasks, i.e., drone-view target localization and drone navigation.
- 456
- 387
- Frequency-Aware Deepfake Detection: Improving Generalizability through Frequency Space Domain Learning369
This research addresses the challenge of developing a universal deepfake detector that can effectively identify unseen deepfake images despite limited training data with a novel frequency-aware approach called FreqNet, centered around frequency domain learning, specifically designed to enhance the generalizability of deepfake detectors.
- 340
- PixelLM: Pixel Reasoning with Large Multimodal Model313
This work introduces PixelLM, an effective and efficient LMM for pixel-level reasoning and understanding, and proposes a target refinement loss to enhance the model's ability to differentiate between multiple targets, leading to substantially improved mask quality.
- Collaborative Video Object Segmentation by Foreground-Background Integration295
This paper proposes Collaborative video object segmentation by Foreground-Background Integration (CFBI) approach, which implicitly imposes the feature embedding from the target foreground object and its corresponding background to be contrastive, promoting the segmentation results accordingly.
- Revisiting RCNN: On Awakening the Classification Power of Faster RCNN285
The potential of detector classification power is demonstrated by a simple, effective, and widely-applicable Decoupled Classification Refinement (DCR) network and failure cases of state-of-the-art detectors are analyzed.
- Forgery-aware Adaptive Transformer for Generalizable Synthetic Image Detection282
This paper develops a forgery-aware adapter to adapt image features to discern and inte-grate local forgery traces within image and frequency do-mains, and finds that considering the contrastive ob-jectives between adapted image features and text prompt embeddings results in a nontrivial generalization improvement.
- 273
- Self-produced Guidance for Weakly-Supervised Object Localization269
Self-produced Guidance (SPG) masks which separate the foreground i.e., the object of interest, from the background to provide the classification networks with spatial correlation information of pixels are proposed.
- Attentive Contexts for Object Detection269
This work develops a novel attention to context convolution neural network (AC-CNN)-based object detection model that effectively incorporates global and local contextual information into the region-based CNN (e.g., fast R-CNN and faster R- CNN) detection framework and provides better object detection performance.
- Devil in the Details: Towards Accurate Single and Multiple Human Parsing266
This paper identifies several useful properties, including feature resolution, global context information and edge details, and performs rigorous analyses to reveal how to leverage them to benefit the human parsing task, resulting in a simple yet effective Context Embedding with Edge Perceiving (CE2P) framework for single human parsing.
- Differential Treatment for Stuff and Things: A Simple Unsupervised Domain Adaptation Method for Semantic Segmentation256
The reason why the current adversarial loss is often unstable in minimizing the distribution discrepancy is revealed and it is shown that the method can help ease this issue by minimizing the most similar stuff and instance features between the source and the target domains.
- Referring Image Segmentation via Cross-Modal Progressive Comprehension250
A Cross-Modal Progressive Comprehension (CMPC) module and a Text-Guided Feature Exchange (TGFE) module to effectively address the challenging task of referring image segmentation.
- Proposal-Free Network for Instance-Level Object Segmentation239
A Proposal-Free Network (PFN) is proposed to address the instance-level object segmentation problem, which outputs the numbers of instances of different categories and the pixel-level information on i) the coordinates of the instance bounding box each pixel belongs to, and ii) the confidences ofDifferent categories for each pixel, based on pixel-to-pixel deep convolutional neural network.
- Agriculture-Vision: A Large Aerial Image Database for Agricultural Pattern Analysis236
This work presents Agriculture-Vision: a large-scale aerial farmland image dataset for semantic segmentation of agricultural patterns, and proposes an effective model designed for aerial agricultural pattern recognition.
- Deep Learning with S-Shaped Rectified Linear Activation Units236
A novel S-shaped rectifiedlinear activation unit (SReLU) to learn both convex and non-convex functions, imitating the multiple function forms given by the two fundamental laws, namely the Webner-Fechner law and the Stevens law, in psychophysics and neural sciences is proposed.
- Reliability Does Matter: An End-to-End Weakly Supervised Semantic Segmentation Approach233
This work harnesses the image-level labels to produce reliable pixel-level annotations and design a fully end-to-end network to learn to predict segmentation maps and gets a new state-of-the-art performance on the Pascal VOC.
- SLCA: Slow Learner with Classifier Alignment for Continual Learning on a Pre-trained Model225
This work presents an extensive analysis for continual learning on a pre-trained model (CLPM), and attributes the key challenge to a progressive overfitting problem, and proposes a simple but extremely effective approach named Slow Learner with Classifier Alignment (SLCA), which further improves the classification layer by modeling the class-wise distributions and aligning the classification layers in a post-hoc fashion.
- Deep Self-Taught Learning for Weakly Supervised Object Localization218
A deep self-taught learning approach is proposed, which makes the detector learn the object-level features reliable for acquiring tight positive samples and afterwards re-train itself based on them, which progressively improves its detection ability and localizes more informative positive samples.
- Cross-Modal Progressive Comprehension for Referring Segmentation202
This paper proposes a cross-modal progressive comprehension (CMPC) scheme to effectively mimic human behaviors and implements it as a CMPC-I (Image) module and aCMPC-V (Video) module to improve referring image and video segmentation models.
- 202
- AlignSeg: Feature-Aligned Segmentation Networks199
This paper proposes Feature-Aligned Segmentation Networks (AlignSeg), a feature aggregation network that adopts a simple learnable interpolation strategy to learn transformation offsets of pixels that can effectively relieve the feature misalignment issue caused by multi-resolution feature aggregation.
- VehicleNet: Learning Robust Visual Representation for Vehicle Re-Identification195
This work proposes to build a unique large-scale vehicle dataset by harnessing four public vehicle datasets, and design a simple yet effective two-stage progressive approach to learning more robust visual representation from VehicleNet.
- L2G: A Simple Local-to-Global Knowledge Transfer Framework for Weakly Supervised Semantic Segmentation191
L2G, a simple online local-to-global knowledge transfer framework for high-quality object attention mining, observes that classification models can discover object regions with more details when replacing the input image with its local patches, and utilizes a global network to learn complementary attention knowledge across multiple local attention maps online.
- Collaborative Video Object Segmentation by Multi-Scale Foreground-Background Integration187
Based on CFBI, a multi-scale matching structure is introduced and an Atrous Matching strategy is proposed, resulting in a more robust and efficient framework, CFBI+.
- IAN: The Individual Aggregation Network for Person Search172
This work proposes a novel Individual Aggregation Network (IAN) that can accurately localize persons by learning to minimize intra-person feature variations and introduces a novel center loss that can increase the intra-class compactness of feature representations.
- Content-Consistent Matching for Domain Adaptive Semantic Segmentation168
This paper considers the adaptation of semantic segmentation from the synthetic source domain to the real target domain and proposes a content-consistent matching (CCM) model, which can alleviate the domain bias caused by those content-irrelevant synthetic images.
- TS$$^{2}$$C: Tight Box Mining with Surrounding Segmentation Context for Weakly Supervised Object Detection163
This work provides a simple approach to discover tight object bounding boxes with only image-level supervision, called Tight box mining with Surrounding Segmentation Context (TS2C), developed based on two key properties of desirable bounding box: high purity and high completeness.
- 161
- 159
- 157
- Affinity Attention Graph Neural Network for Weakly Supervised Semantic Segmentation156
This paper proposes affinity attention graph neural network, in which an affinity attention layer is designed to acquire the short- and long- distance information from soft graph edges to accurately propagate semantic labels from the confident seeds to the unlabeled pixels.
- SPGNet: Semantic Prediction Guidance for Scene Parsing148
By carefully re-weighting features across stages, a two-stage encoder-decoder network coupled with the proposed Semantic Prediction Guidance (SPG) module can significantly outperform its one-stage counterpart with similar parameters and computations.
- 4DGen: Grounded 4D Content Generation with Spatial-temporal Consistency145
Compared to previous image-to-4D and text-to-4D works, 4DGen supports grounded generation, offering users enhanced control and improved motion generation capabilities, a feature difficult to achieve with previous methods.
- 140
- VisDrone-DET2018: The Vision Meets Drone Object Detection in Image Challenge Results139
A large-scale drone-based dataset, including 8, 599 images with rich annotations, including object bounding boxes, object categories, occlusion, truncation ratios, etc, is released, to narrow the gap between current object detection performance and the real-world requirements.
- 136
- Contrastive Adaptation Network for Single- and Multi-Source Domain Adaptation132
This paper proposes contrastive adaptation network (CAN) that optimizes a new metric named Contrastive Domain Discrepancy explicitly modeling the intra-classdomain discrepancy and the inter-class domain discrepancy and develops class-aware sampling to enable more efficient and effective training.
- 126
- Inter-Image Communication for Weakly Supervised Localization115
This paper proposes to leverage pixel-level similarities across different objects for learning more accurate object locations in a complementary way, and proposes two kinds of constraints that can benefit each other to learn consistent pixel- level features within the same categories, and improve the quality of localization maps.
- Learning to segment with image-level annotations110
This paper explores a more challenging problem by learning to segment under image-level annotations by incorporating the hypotheses-aware classification and cross-image contextual refinement in a supervised manner, and proposes two network training strategies for achieving good segmentation performance.
- 110
- 108
- Boundary Perception Guidance: A Scribble-Supervised Semantic Segmentation Approach99
This work presents a novel Boundary Perception Guidance approach, which consists of two basic components, i.e. prediction refinement and boundary regression, and employs class-agnostic edge maps for supervision to effectively guide the segmentation network in localizing the boundaries between different semantic regions.
- CLE Diffusion: Controllable Light Enhancement Diffusion Model97
Controllable Light Enhancement Diffusion Model, dubbed CLE Diffusion, a novel diffusion framework to provide users with rich controllability, is built with a conditional diffusion model and an illumination embedding is introduced to let users control their desired brightness level.
- Memory Aggregation Networks for Efficient Interactive Video Object Segmentation95
The proposed MA-Net integrates the interaction and the propagation operations into a single network, which significantly promotes the efficiency of iVOS in the scheme of multi-round interactions, and proposes a simple yet effective memory aggregation mechanism to record the informative knowledge from the previous interaction rounds, improving the robustness in discovering challenging objects of interest greatly.
- Product1M: Towards Weakly Supervised Instance-Level Product Retrieval via Cross-Modal Pretraining91
A novel model named Cross-modal contrAstive Product Transformer for instance-level prodUct REtrieval (CAPTURE) is proposed, that excels in capturing the potential synergy between multi- modal inputs via a hybrid-stream transformer in a self-supervised manner.
- Modality-Dependent Cross-Media Retrieval91
This article proposes a modality-dependent cross-media retrieval (MDCR) model, where two couples of projections are learned for different cross- media retrieval tasks instead of one couple of projections, and achieves a new state-of-the-art performance on the Wikipedia dataset.
- Learning Mask-aware CLIP Representations for Zero-Shot Segmentation89
Mask-aware loss and self-distillation loss are designed to fine-tune IP-CLIP Encoder, ensuring CLIP is responsive to different mask proposals while not sacrificing transferability, and can be easily learned to make the true positives stand out.
- Left-Right Comparative Recurrent Model for Stereo Matching89
A novel left- right comparative recurrent model to perform left-right consistency checking jointly with disparity estimation and a soft attention mechanism is introduced, which employs the learned error maps for better guiding the model to selectively focus on refining the unreliable regions at the next recurrent step.
- Consistent Structural Relation Learning for Zero-Shot Segmentation.88
This work proposes a Consistent Structural Relation Learning (CSRL) approach to constrain the generating of unseen visual features by exploiting the structural relations between seen and unseen categories and outperforms existing state-of-the-art methods by a large margin.
- Weakly Supervised Scene Parsing with Point-Based Distance Metric Learning84
This paper proposes a Point-based Distance Metric Learning (PDML), which leverages semantic relationship among the annotated points by encouraging the feature representations of the intra- and inter-category points to keep consistent.
- 82
- 81
- 79
- AppAgent v2: Advanced Agent for Flexible Mobile Interactions76
This work introduces a novel LLM-based multimodal agent framework for mobile devices, capable of navigating mobile devices, that emulates human-like interactions and constructs a flexible action space that enhances adaptability across various applications including parser, text and vision descriptions.
- Progressive Semantic-Visual Mutual Adaption for Generalized Zero-Shot Learning73
A debiasing loss is proposed to pursue response consistency between seen and unseen predictions, and a semantic-motivated instance decoder strengthens accurate cross-domain interactions between the matched pair for semantic-related instance adaption, en-couraging the generation of unambiguous visual representations.
- 72
- Frozen CLIP: A Strong Backbone for Weakly Supervised Semantic Segmentation67
The frozen CLIP model is applied as the backbone for semantic feature extraction, and a new decoder is designed to interpret extracted semantic features for final prediction and a refinement module (RFM) is proposed to rectify them dynamically.
- Alleviating Semantic-level Shift: A Semi-supervised Domain Adaptation Method for Semantic Segmentation67
A semi-supervised approach named Alleviating Semantic-level Shift (ASS) is proposed, which can promote the distribution consistency from both global and local views and can beat the oracle model trained on the whole target dataset.
- Global Knowledge Calibration for Fast Open-Vocabulary Segmentation64
A core idea of preserving the generalizable representation when fine-tuning on known classes is proposed, and a text diversification strategy that generates a set of synonyms for each training category is introduced, which prevents the learned representation from collapsing onto specific known category names.
- Multistage Object Detection With Group Recursive Learning61
A new EM-like group recursive learning approach to iteratively refine object proposals by incorporating such context of surrounding proposals and provide an optimal spatial configuration of object detections is proposed.
- Reversible Recursive Instance-Level Object Segmentation61
This work proposes a novel Reversible Recursive Instance-level Object Segmentation (R2-IOS) framework that consists of a reversible proposal refinement sub-network that predicts bounding box offsets for refining the object proposal locations, and an instance-level segmentation sub- network that generates the foreground mask of the dominant object instance in each proposal.
- 58
- SegRefiner: Towards Model-Agnostic Segmentation Refinement with Discrete Diffusion Process57
A model-agnostic solution called SegRefiner is proposed, which offers a novel perspective on segmentation refinement as a data generation process by interpreting segmentation refinement as a data generation process, and which outperforms previous model-agnostic refinement methods by a significant margin.
- 57
- M5Product: Self-harmonized Contrastive Learning for E-commercial Multi-modal Pretraining56
Self-harmonized ContrAstive LEarning (SCALE), a novel pretraining framework that integrates the different modalities into a unified model through an adaptive feature fusion mechanism, is proposed, providing insights into the importance of dataset scale and diversity.
- VideoWorld: Exploring Knowledge Learning from Unlabeled Videos55
This study develops and introduces the Latent Dynamics Model (LDM) as a key component of VideoWorld, an auto-regressive video generation model trained on unlabeled video data, and tests its knowledge acquisition abilities in video-based Go and robotic control tasks.
- Mining Unseen Classes via Regional Objectness: A Simple Baseline for Incremental Segmentation55
The proposed MicroSeg is based on the assumption that background regions with strong objectness possibly belong to those concepts in the historical or future stages, and first splits the given image into hundreds of segment proposals with a proposal generator, to avoid forgetting old knowledge at the current training stage.
- Mask Matching Transformer for Few-Shot Segmentation55
The proposed Mask Matching Transformer (MM-Former), a new paradigm for the few-shot segmentation task, uses a class-agnostic segmenter to decompose the query image into multiple segment proposals and applies a simple matching mechanism to merge the related segment proposals into the final mask guided by the support images.
- 52
- 50
- Scalable Video Object Segmentation With Identification Mechanism47
Two innovative approaches to scalable and effective multi-object modeling for semi-supervised Video Object Segmentation (VOS) are presented, Associating Objects with Transformers (AOT) and Associating Objects with Scalable Transformers (AOST), which display exceptional efficiency and scalability consistently across all six benchmarks.
- 46
- CTP: Towards Vision-Language Continual Pretraining via Compatible Momentum Contrast and Topology Preservation46
A new algorithm: compatible momentum contrast with Topology Preservation, dubbed CTP is proposed, which absorbs the knowledge of the current and previous-task models to flexibly update the modal feature and transfers the knowledge of embedding across tasks while preserving the flexibility of feature adjustment.
- 43
- 43
- 42
- Frequency-Aware Deepfake Detection: Improving Generalizability through Frequency Space Learning41
This research addresses the challenge of developing a universal deepfake detector that can effectively identify unseen deepfake images despite limited training data by introducing a novel frequency-aware approach called FreqNet, centered around frequency domain learning, specifically designed to enhance the generalizability of deepfake detectors.
- Bottom-Up Top-Down Cues for Weakly-Supervised Semantic Segmentation39
This work considers the task of learning a classifier for semantic segmentation using weak supervision in the form of image labels specifying objects present in the image, and adopts an Expectation-Maximization (EM) based approach, which outperforms the state-of-the-art method with a very high margin.
- Modeling the Label Distributions for Weakly-Supervised Semantic Segmentation37
A unified WSSS framework named Adaptive Gaussian Mixtures Model is developed, which leverages a GMM to generate high-quality pseudo labels for more reliable supervision and outperform the state-of-the-art methods under all settings.
- Credible Dual-Expert Learning for Weakly Supervised Semantic Segmentation37
This paper proposes a credible dual-expert learning (CDL) framework to mitigate the noise of pseudo labels, and designs a dual-expert structure to mine credible predictions, which are then processed by the noise correction module to update pseudo labels in an online way.
- StableLLaVA: Enhanced Visual Instruction Tuning with Synthesized Image-Dialogue Data36
This work proposes a novel data collection methodology that synchronously synthesizes images and dialogues for visual instruction tuning, marrying the abilities of ChatGPT and text-to-image generative models to yield a diverse and controllable dataset with varied image content.
- Introduction to the Special Issue on Trustworthy Multimedia Computing and Applications in Urban Scenes36
The goal of this special issue (SI) is to encourage a collaborative endeavor to explore the possibilities and obstacles associated with trustworthy multimedia analysis and gather inventive approaches, systems, and applications focused on trustworthy multimedia computing and its applications in urban environments.
- VMFormer: End-to-End Video Matting with Transformer35
VMFormer is proposed: a transformer-based end-to-end method for video matting that leverages self-attention layers to build global integration of feature sequences with short-range temporal modeling on successive frames and outperforms previous CNN-based video matting methods on synthetic benchmarks.
- 34
- Unified Frequency-Assisted Transformer Framework for Detecting and Grounding Multi-modal Manipulation34
The Unified Frequency-Assisted transFormer framework, named UFAFormer, is presented, which comprises two symmetric cross-modal interaction modules responsible for gathering modality-specific forgery information, along with a fusing interaction module for aggregation of both modalities.
- AMC: Adaptive Multi-expert Collaborative Network for Text-guided Image Retrieval32
Extensive experiments demonstrate that due to benefits from the image-text composing representation produced by an adaptive multi-expert collaboration mechanism, AMC has better retrieval performance and zero-shot generalization ability than the state-of-the-art method while keeping the lightweight model and fast retrieval speed.
- 32
- 31
- Sketch-Guided Scenery Image Outpainting31
This work proposes an encoder-decoder based network to conduct sketch-guided outpainting, where two alignment modules are adopted to impose the generated content to be realistic and consistent with the provided sketches.
- 31
- Topographic NMF for Data Representation30
A topographic NMF (TNMF), which imposes a topographic constraint on the encoding factor as a regularizer during matrix factorization, and which will force the encodings to be organized in a topographical map so that the feature invariance can be promoted.
- Foundations and Recent Trends in Multimodal Mobile Agents: A Survey29
This survey provides a comprehensive review of mobile agent technologies, focusing on recent advancements that enhance real-time adaptability and multimodal interaction, and categorizes these advancements into two main approaches: prompt-based methods, which utilize large language models for instruction-based task execution, and training-based methods, which fine-tune multimodal models for mobile-specific applications.
- Multi-human Parsing with a Graph-based Generative Adversarial Model28
A novel multi- human parsing model named MH-Parser is introduced, which uses a graph-based generative adversarial model to address the challenges of close-person interaction and occlusion in multi-human parsing.
- Deep Attribute-preserving Metric Learning for Natural Language Object Retrieval28
A deep Attribute-Preserving Metric learning (AP-Metric) framework that jointly generates unique query-sensitive region proposals and conducts novel cross-modal feature learning that explicitly pursues consistency over semantic attribute abstraction within both domains for deep metric learning is proposed.
- UAVM '23: 2023 Workshop on UAVs in Multimedia: Capturing the World from a New Perspective27
This workshop covers various topics related to UAV multimedia, including aerial image and video processing, machine learning for UAV data analysis, UAV swarm technology, and UAV-based multimedia applications.
- The 1st Agriculture-Vision Challenge: Methods and Results27
The first Agriculture-Vision Challenge aims to encourage research in developing novel and effective algorithms for agricultural pattern recognition from aerial images, especially for the semantic segmentation task associated with the challenge dataset.
- Transferable Semi-Supervised Semantic Segmentation27
A novel transferable semi-supervised semantic segmentation model that can transfer the learned segmentation knowledge from a few strong categories with pixel- level annotations to unseen weak categories with only image-level annotations is proposed, significantly broadening the applicable territory of deep segmentation models.
- 26
- 26
- 25
- 25
- 25
- 25
- Understanding and Accelerating Neural Architecture Search With Training-Free and Theory-Grounded Metrics24
This work designs a principled and unified training-free framework for Neural Architecture Search (NAS), with high performance, low cost, and in-depth interpretation, and presents a unified framework to understand and accelerate NAS by disentangling “TEG” characteristics of searched networks – Trainability, Expressivity, Generalization – all assessed in a training-free manner.
- 24
- Learning to Segment Human by Watching YouTube24
A very-weakly supervised learning framework for human segmentation task, where only an imperfect human detector is available along with massive weakly-labeled YouTube videos, and reaches a new state-of-the-art performance on the human segmentsation task.
- 23
- Proximal Iteratively Reweighted Algorithm with Multiple Splitting for Nonconvex Sparsity Optimization23
This paper proposes the Proximal Iteratively REweighted (PIRE) algorithm for solving a general problem, which involves a large body of nonconvex sparse and structured sparse related problems and proves that the proposed methods converge and any limit solution is a stationary point.
- RIO: A Benchmark for Reasoning Intention-Oriented Objects in Open Environments22
The ability of some existing models to reason intention-oriented objects in open environments is evaluated and the proposed RIO is evaluated, specifically designed to incorporate diverse real-world scenarios and a wide range of object categories.
- 22
- 22
- 22
- 22
- Data-Independent Operator: A Training-Free Artifact Representation Extractor for Generalizable Deepfake Detection21
This work shows that the small and training-free filter is sufficient to capture more general artifact representations and defines the DIO as Data-Independent Operator (DIO) to achieve appealing improvements on unseen sources.
- Intent3D: 3D Object Detection in RGB-D Scans Based on Human Intention20
IntentNet, the unique approach, designed to tackle this intention-based detection problem, focuses on three key aspects: intention understanding, reasoning to identify object candidates, and cascaded adaptive learning that leverages the intrinsic priority logic of different losses for multiple objective optimization.
- 20
- 20
- Towards Real World Human Parsing: Multiple-Human Parsing in the Wild.20
A novel Multiple-Human Parser (MH-Parser) is proposed, which considers both the global context and local cues for each person in the parsing process and is demonstrated to outperform the naive "detect-and-parse" approach by a large margin.
- Kernel Reconstruction ICA for Sparse Representation20
A kernel RICA (kRICA) model is proposed to nonlinearly capture sparse representation in feature space and is extended to a supervised one by introducing a class-driven discrimination constraint, such that the data samples from the same class are well represented on the basis of the corresponding subset of basis vectors.
- 19
- Cylin-Painting: Seamless 360° Panoramic Image Outpainting and Beyond19
A deep analysis of the differences between inpainting and outpainting is provided, which essentially depends on how the source pixels contribute to the unknown regions under different spatial arrangements, and a learnable positional embedding strategy to incorporate the missing component of positional encoding into the cylinder convolution, which significantly improves the panoramic results.
- 19
- 19
- Temporal Spiking Recurrent Neural Network for Action Recognition19
The proposed TSRNN employs a novel spiking architecture which utilizes the local discriminative features from high-confidence reliable frames as spiking signals which can guide RNN to correct its long-term memory across multiple frames from contamination caused by noisy frames with distracting factors.
- 18
- 18
- Dual-view X-ray Detection: Can AI Detect Prohibited Items from Dual-view X-ray Images like Humans?17
The Auxiliary-view Enhanced Network (AENet), a novel detection framework that leverages both the main and auxiliary views of the same object, is proposed, which exhibits strong generalization across seven different detection models for X-ray Inspection.
- 17
- SpeechR: A Benchmark for Speech Reasoning in Large Audio-Language Models16
SpeechR, a unified benchmark for evaluating reasoning over speech in large audio-language models, establishes a structured benchmark for evaluating reasoning in spoken language, enabling more targeted analysis of model capabilities across diverse dialogue-based tasks.
- Learning Hierarchical Color Guidance for Depth Map Super-Resolution16
This article rethink the utilization of color information and proposes a hierarchical color guidance network (HCGNet) to achieve DSR, which achieves more competitive performance both qualitatively and quantitatively.
- 16
- Focus Longer to See Better: Recursively Refined Attention for Fine-Grained Image Classification16
This paper shows through interpretability techniques how the network focus changes from coarse to fine details, and shows that a simple attention model can aggregate (weighted) these finer details to focus on the most dominant discriminative part of the image.
- 15
- 15
- CLIP-GS: Unifying Vision-Language Representation with 3D Gaussian Splatting14
This paper introduces the GS Tokenizer to generate serialized gaussian tokens, which are then processed through transformer layers pre-initialized with weights from point cloud models, resulting in the 3DGS embeddings, and introduces an image voting loss to guide the directionality and convergence of gradient optimization.
- PSVMA+: Exploring Multi-Granularity Semantic-Visual Adaption for Generalized Zero-Shot Learning14
A multi-granularity progressive semantic-visual mutual adaption (PSVMA+) network, where sufficient visual elements across granularity levels can be gathered to remedy the granularity inconsistency, and experimental results demonstrate that PSVMA+ consistently outperforms state-of-the-art methods.
- 14
- Single-Stage Multi-human Parsing via Point Sets and Center-based Offsets14
A high-performance Single-stage Multi-human Parsing (SMP) deep architecture that decouples the multi-human parsing problem into two fine-grained sub-problems,i.e., locating the human body and parts.
- Entity-Graph Enhanced Cross-Modal Pretraining for Instance-level Product Retrieval14
A novel Entity-Graph Enhanced Cross-Modal Pretraining (EGE-CMP) model is proposed for instance-level commodity retrieval, which explicitly injects entity knowledge in both node-based and subgraph-based ways into the multi-modal networks via a self-supervised hybrid-stream transformer.
- 14
- Class-Driven Non-Negative Matrix Factorization for Image Representation14
A semi-supervised class-driven NMF method to associate a class label with each basis vector by introducing an inhomogeneous representation cost constraint so that the discriminability in new representations could be boosted.
- DCEdit: Dual-Level Controlled Image Editing via Precisely Localized Semantics13
A Precise Semantic Localization strategy that leverages visual and textual self-attention to enhance the cross-attention map, which can serve as a regional cues to improve editing performance, and a Dual-Level Control mechanism for incorporating regional cues at both feature and latent levels.
- UniParser: Multi-Human Parsing With Unified Correlation Representation Learning13
By unifying instance-level and category-level output, UniParser circumvents manually designed post-processing techniques and surpasses state-of-the-art methods, achieving 49.3% AP on MHPv2.0 and 60.4% AP on CIHP.
- 13
- 13
- 13
- Mining Pixels: Weakly Supervised Semantic Segmentation Using Image Labels13
This work considers the task of learning a classifier for semantic segmentation using weak supervision, in this case, image labels specifying the objects within the image, and proposes minimizing the combination of the standard softmax loss and the KL divergence between the true latent posterior and the likelihood given by the CNN.
- CoMBO: Conflict Mitigation via Branched Optimization for Class Incremental Segmentation12
This work introduces a novel approach, Conflict Mitigation via Branched Optimization (CoMBO), designed to explicitly refine queries for new classes through lightweight, class-specific adapters, and develops two strategies to further mitigate the conflict following the branched structure.
- ReGO: Reference-Guided Outpainting for Scenery Image12
ReGO (Reference-Guided Outpainting), a new method for the task of sketch-guided image outpainting that leverages neighboring reference images to synthesize texture-rich results by transferring pixels from them to facilitate texture compensating of the target image.
- 12
- 12
- 12
- Segment Anything with Precise Interaction11
Pi-SAM possesses a strong perception capability for the extremely fine details in high-resolution images, enabling it to generate high-precision segmentation masks and ensures highly efficient model fine-tuning to achieve the above performance.
- SLCA++: Unleash the Power of Sequential Fine-tuning for Continual Learning with Pre-training11
This work presents an in-depth analysis of the progressive overfitting problem from the lens of Seq FT, and introduces the advanced Slow Learner with Classifier Alignment (SLCA++) framework to unleash the power of Seq FT, serving as a strong baseline approach for CLPT.
- 11
- LEDNet: a multimodal foundation model for robust deepfake detection10
This work is the first to exploit multi-modal technologies for generalized deepfake detection through an un-trained-backbone mechanism, which utilizes pre-trained vision-language models without the need for additional retraining.
- Enhanced Visual Instruction Tuning with Synthesized Image-Dialogue Data9
This work proposes a novel methodology for data collection based on annotations derived from benchmark datasets to construct image-dialogue datasets for training purposes, akin to instruction tuning in LLMs.
- ACTRESS: Active Retraining for Semi-supervised Visual Grounding9
The ACTive REtraining approach for Semi-Supervised Visual Grounding, abbreviated as ACTRESS, consists of an active sampling strategy and a selective retraining strategy that retrains the model with periodic re-initialization of specific parameters, facilitating the model's escape from local minima.
- 9
- 9
- Clicking Matters: Towards Interactive Human Parsing9
This work focuses on Interactive Human Parsing, which aims to segment a human image into multiple human body parts with guidance from users’ interactions and proposes a semantic-perceiving loss to augment the training, which can effectively exploit the semantic relationships of clicks for better optimization.
- 9
- 9
- 9
- Progressive Neighborhood Aggregation for Semantic Segmentation Refinement8
A progressive neighborhood aggregation (PNA) framework to refine the semantic segmentation prediction, resulting in an end-to-end solution that can perform the coarse prediction and refinement in a unified network.
- Slim Scissors: Segmenting Thin Object from Synthetic Background8
Slim Scissors is developed, which enables quick extraction of elon-gated thin parts by simply brushing some coarse scribbles and is model-agnostic and seamlessly applicable to existing state-of-the-art interactive segmentation models.
- 7
- Who Can Withstand Chat-Audio Attacks? An Evaluation Benchmark for Large Audio-Language Models7
This paper evaluates six state-of-the-art LALMs with voice interaction capabilities, including Gemini-1.5-Pro, GPT-4o, and others, using three distinct evaluation methods on the CAA benchmark, demonstrating that GPT-4o exhibits the highest level of resilience.
- AdGPT: Explore Meaningful Advertising with ChatGPT7
AdGPT is introduced, a model that leverages visual expert analysis to guide Large Language Models (LLMs) in generating adaptive reasoning chains and can intelligently comprehend meaningful ads regarding category, content, and sentiment.
- ForenX: Towards Explainable AI-Generated Image Detection with Multimodal Large Language Models7
ForenX is presented, a novel method that not only identifies the authenticity of images but also provides explanations that resonate with human thoughts that enhance the generalization of forgery detection and provide explanations that are accurate, relevant, and comprehensive.
- PanoWorld-X: Generating Explorable Panoramic Worlds via Sphere-Aware Video Diffusion7
This work proposes PanoWorld-X, a novel framework for high-fidelity and controllable panoramic video generation with diverse camera trajectories, and introduces a Sphere-Aware Diffusion Transformer architecture that reprojects equirectangular features onto the spherical surface to model geometric adjacency in latent space, significantly enhancing visual fidelity and spatiotemporal continuity.
- 7
- 7
- Magic-wall7
An edge-aware fully convolutional neural network (FCN) is proposed for indoor semantic scene parsing, in which a novel edge-prior branch is introduced to better identify the boundary of different semantic regions.
- 7
- TiP4GEN: Text to Immersive Panorama 4D Scene Generation6
TiP4GEN is introduced, an advanced text-to-dynamic panorama scene generation framework that enables fine-grained content control and synthesizes motion-rich, geometry-consistent panoramic 4D scenes.
- 6
- Pay Less Attention to Deceptive Artifacts: Robust Detection of Compressed Deepfakes on Online Social Networks6
PLADA is proposed, a novel framework designed to tackle the lack of paired data and the ineffective use of compressed images in deepfake detection, and introduces the ``block effect" as a critical factor in deepfake detection, providing a robust solution for open-world scenarios.
- Attend and Enrich: Enhanced Visual Prompt for Zero-Shot Learning6
This work proposes AENet, which endows semantic information into the visual prompt to distill semantic-enhanced prompt for visual representation enrichment, enabling effective knowledge transfer for ZSL.
- ROSE: Revolutionizing Open-Set Dense Segmentation with Patch-Wise Perceptual Large Multimodal Model6
This work proposes ROSE, a Revolutionary Open-set dense SEgmentation LMM, which enables dense mask prediction and open-category generation through patch-wise perception, and introduces a conversation-based refinement paradigm, integrating the prediction result from previous step with textual prompt for revision.
- 6
- 6
- 6
- Revisiting Pre-training: An Efficient Training Method for Image Classification6
This work proposes to continually repeat a Drop-and-Pick (DaP) learning strategy, which can be leveraged to train classification networks in a more efficient way and is equipped with strong knowledge transferability when used for downstream tasks, especially for hard cases.
- Data-Efficient Inference of Neural Fluid Fields via SciML Foundation Model5
This work demonstrates that SciML foundation models can significantly reduce the data costs of inferring real-world 3D fluid dynamics with improved generalization, and showcases the practical applicability of SciML foundation models in real-world fluid dynamics.
- 5
- Omni-supervised Facial Expression Recognition: A Simple Baseline.5
This work proposes to perform omni-supervised learning by exploiting the labeled samples together with a large number of unlabeled data to enhance the robustness of the learned models for various scenarios and adopts a dataset distillation strategy to distill the target task-related knowledge from the new mined samples and compressed them into a very small set of images.
- Learning Trimaps via Clicks for Image Matting4
This work introduces Click2Trimap, an interactive model that is capable of predicting high-quality trimaps and alpha mattes with minimal user click inputs and proposes a powerful iterative three-class training strategy and a dedicated simulation function, making Click2Trimap exhibit versatility across various scenarios.
- Collaborative Feature-Logits Contrastive Learning for Open-Set Semi-Supervised Object Detection4
This work introduces a feature-level clustering method using contrastive loss to clarify vector boundaries in the feature space and highlight class differences in semi-supervised object detection, and enhances its ability to effectively distinguish between ID and OOD classes.
- Disentangled Pre-training for Image Matting4
Extensive experimental evaluations show that the proposed approach outperforms both the state-of-the-art matting methods and other alternative self-supervised initialisation approaches by a large margin, and shows the robustness of the proposed approach over different backbone architectures.
- 4
- 4
- FisheyeEX: Polar Outpainting for Extending the FoV of Fisheye Lens4
A FisheyeEX method that extends the FoV of the fisheYE lens by outpainting the invalid regions, improving the integrity of captured scenes and significantly outperforms the state-of-the-art methods.
- One Shot Domain Adaptation for Person Re-Identification4
This work introduces a similarity-guided strategy to progressively assign pseudo labels to unlabeled instances with different confidence scores, which are in turn leveraged as weights to guide the optimization as training goes on and makes significant improvement in the domain adaptation tasks of re-ID.
- Accumulated reconstruction error vector (AREV): a semantic representation for cross-media retrieval4
A novel semantic representation of cross-media, called accumulated reconstruction error vector (AREV), is proposed, which includes category-specific dictionary learning, media sample reconstruction, and accumulative reconstruction error concatenation.
- 3
- Transferable and Principled Efficiency for Open-Vocabulary Segmentation3
The core strategy is to make the efficiency principled and thus seamlessly transferable from one OVS framework to others without further customization, by utilizing smaller models that incur lower training costs.
- 3
- One-shot In-context Part Segmentation3
This framework offers a novel approach to part segmentation that is training-free, flexible, and data-efficient, requiring only a single in-context example for precise segmentation with superior generalization ability.
- 3
- 3
- 3
- Taming Generative Synthetic Data for X-Ray Prohibited Item Detection2
A one-stage X-ray security image synthesis pipeline (Xsyn) based on text-to-image generation, which incorporates two effective strategies to improve the usability of synthetic images and is the first to achieve high-quality X-ray security image synthesis without extra labor cost.
- CoCoDiff: Correspondence-Consistent Diffusion Model for Fine-grained Style Transfer2
CoDiff is proposed, a novel training-free and low-cost style transfer framework that leverages pretrained latent diffusion models to achieve fine-grained, semantically consistent stylization.
- 2
- AlignGen: Boosting Personalized Image Generation with Cross-Modality Prior Alignment2
AlignGen, a Cross-Modality Prior Alignment mechanism that enhances personalized image generation by introducing a learnable token to bridge the gap between the textual and visual priors, and employing a selective cross-modal attention mask within the multi-modal attention mechanism to further align the priors is proposed.
- IPSeg: Image Posterior Mitigates Semantic Drift in Class-Incremental Segmentation2
IPSeg introduces two key mechanisms: leveraging image posterior probabilities to align optimization across stages and mitigate the effects of separate optimization, and employing semantics decoupling to handle noisy semantics and tailor learning strategies for different semantics.
- 2
- Long-range diffusion for weakly camouflaged object segmentation2
The long-range diffusion network (LRDNet) is proposed to diffuse the sparse annotations for improving WSCOS performance and a novel gated local saliency coherence (GLSC) loss is designed to efficiently diffuse limited annotation information across the entire image to supplement the supervision by the unidirectional gating.
- 2
- Behavior Backdoor for Deep Learning Models2
This study proposes the first pipeline of implementing behavior backdoor, i.e., the Quantification Backdoor (QB) attack, upon exploiting model quantification method as the set trigger, and introduces the behavior-driven backdoor object optimizing method by a bi-target behavior backdoor training loss, thus it could guide the poisoned model optimization direction.
- 2
- 2
- 2
- 2
- 2
- 2
- 2
- Heterogeneous data alignment for cross-media computing2
This paper gives a short introduction about heterogeneous data alignment, and two ongoing works in the group about heterogeneity data alignment are discussed, i.e., consistent pattern mining and modality-dependent cross-media retrieval.
- A Simple Switchable Framework for Open-Vocabulary Video Instance Segmentation1
This work designs a simple yet effective switchable open-vocabulary VIS (SOV) framework that outperforms previous methods by large margins on various benchmarks while maintaining faster inference speeds and proposes a progressive “Long-Image, Short-Video” training pipeline.
- 1
- 1
- StructGen: Disambiguating Multi-Reference Image Generation via Structured Context Modeling1
StructGen is proposed, which employs a structured, dictionary-like format to encode multiple reference images, thereby enabling explicit and unambiguous specification of generation intentions, and consistently outperforms existing methods on both semantic alignment and detailed reference-generation consistency.
- 1
- 1
- 1
- 1
- LayerMatch: Do Pseudo-labels Benefit All Layers?1
LayerMatch, which integrates two layer-specific pseudo-label strategies, can avoid the severe interference of noisy pseudo-labels in the linear classification layer while accelerating the clustering capability of the feature extraction layer.
- 1
- 1
- Adversarial task-specific learning1
A new Adversarial Task-Specific Learning approach to learn distinct embeddings for different retrieval tasks, i.e. images retrieve texts (I2T) or texts retrieve images (T2I), embedded in a deep neural network, which can be learned in an end-to-end manner.
- 1
- –
- –
- OVFormer+: Improved Open-Vocabulary Video Instance Segmentation via Text-Guided Unified Embedding Alignment–
This work designs and trains a novel OV-VIS baseline called OVFormer+.
- Revisiting Pixel Erasure: A Simple Method for Weakly Supervised Video Instance Segmentation–
It is shown that VIS without video mask annotations can perform comparably to or even outperform state-of-the-art fully supervised methods on YouTube-VIS (YTVIS) 2019 and 2021 and is demonstrated that the pixel erasure strategy can also advance other weakly supervised tasks that involve learning from pseudo masks.
- –
- Entropy Regularized Simple Multiple Kernel K-Means with Adaptive Deviation Correction–
This paper proposes an entropy regularized simple multiple kernel k-means with adaptive deviation correction (ERSMKKM-ADC), which incorporates an entropy regularization term to balance the distribution of multiple kernel weights, preventing them from becoming overly sparse or excessively concentrated.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- The SkatingVerse Workshop & Challenge: Methods and Results–
A brief summary of the SkatingVerse Workshop&Challenge is provided including brief introductions to the top three methods including brief introductions to the top three methods.
- BiEPNet: Bilateral Edge-perceiving Network for High-Resolution Human Parsing–
The Bilateral Edge-Perceiving Network (BiEPNet), consisting of a resources-friendly semantic-perceiving branch to acquire sufficient global information and a simple yet effective edge-perceiving branch used to refine details is proposed, which effectively outperforms the state-of-the-art methods.
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Study on the Improvement of Traffic Congestion by Community Open–
The paper establishes a road evaluation system based on congestion degree and vehicle buffering, and establishes the vehicle traffic model of neighborhood road, and the relationship between the area S of the open area and the suitable open level n.
- Structured Deep Learning for Pixel-level Understanding–
This tutorial reviews recent progresses for pixel-level understanding with structured deep learning, including 1) human-centric analysis: human parsing and pose estimation; 2) part-based analysis: part and face parsing; 3) weakly-supervised analysis: object localization and semantic segmentation; 4) depth estimation: stereo matching.
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.