Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks137 from public data
- Coordinate Attention for Efficient Mobile Network Design6,424
This paper proposes a novel attention mechanism for mobile networks by embedding positional information into channel attention, which it is called "coordinate attention", which factorizes channel attention into two 1D feature encoding processes that aggregate features along the two spatial directions.
- Deeply Supervised Salient Object Detection with Short Connections1,458
This paper proposes a new salient object detection method by introducing short connections to the skip-layer structures within the HED architecture, which takes full advantage of multi-level and multi-scale features extracted from FCNs, providing more advanced representations at each layer, a property that is critically needed to perform segment detection.
- Rotate to Attend: Convolutional Triplet Attention Module1,142
The method is simple as well as efficient and can be easily plugged into classic backbone networks as an add-on module and supports the intuition on the importance of capturing dependencies across dimensions when computing attention weights.
- A Simple Pooling-Based Design for Real-Time Salient Object Detection1,100
This work solves the problem of salient object detection by investigating how to expand the role of pooling in convolutional neural networks by building a global guidance module (GGM) and designing a feature aggregation module (FAM) to make the coarse-level semantic information well fused with the fine-level features from the top-down path- way.
- 939
- Large Selective Kernel Network for Remote Sensing Object Detection887
This paper considers these priors and proposes the lightweight Large Selective Kernel Network (LSKNet), which can dynamically adjust its large spatial receptive field to better model the ranging context of various objects in remote sensing scenarios.
- Strip Pooling: Rethinking Spatial Pooling for Scene Parsing817
Beyond conventional spatial pooling that usually has a regular shape of NxN, a new pooling strategy is introduced, called strip pooling, which considers a long but narrow kernel, i.e., 1xN or Nx1.
- 653
- 495
- VOLO: Vision Outlooker for Visual Recognition422
A new simple and generic architecture, termed Vision Outlooker (VOLO), which implements a novel outlook attention operation that dynamically conduct the local feature aggregation mechanism in a sliding window manner across the input image, which can more efficiently encode fine-level features that are essential for high-performance visual recognition.
- 373
- Salient Objects in Clutter: Bringing Salient Object Detection to the Foreground346
This work identifies a serious design bias of existing SOD datasets which assumes that each image contains at least one clearly outstanding salient object in low clutter, and proposes a new high quality dataset and updates the previous saliency benchmark.
- 334
- Conv2Former: A Simple Transformer-Style ConvNet for Visual Recognition290
A deep look at the internal structure of self-attention is taken and a simple Transformer style convolutional neural network (ConvNet) is presented for visual recognition and a family of hierarchical ConvNets using the proposed convolutional modulation, termed Conv2Former is built.
- Rethinking Bottleneck Structure for Efficient Mobile Network Design275
This paper proposes to flip the structure and present a novel bottleneck design, called the sandglass block, that performs identity mapping and spatial transformation at higher dimensions and thus alleviates information loss and gradient confusion effectively and adds it into the search space of neural architecture search method DARTS.
- 260
- 254
- YOLO-MS: Rethinking Multi-Scale Representation Learning for Real-Time Object Detection246
The core design is based on a series of investigations on how multi-branch features of the basic block and convolutions with different kernel sizes affect the detection performance of objects at different scales, and the outcome is a new strategy that can significantly enhance multi-scale feature representations of real-time object detectors.
- LSKNet: A Foundation Lightweight Backbone for Remote Sensing237
This paper considers the significance of the identified priors and proposes a lightweight Large Selective Kernel Network (LSKNet) backbone, which can dynamically adjust its large spatial receptive field to better model the ranging context of various objects in remote sensing scenarios.
- AMT: All-Pairs Multi-Field Transforms for Efficient Frame Interpolation229
This work presents All-Pairs Multi-Field Transforms (AMT), a new network architecture for video frame interpolation based on two essential designs that derive multiple groups of fine-grained flow fields from one pair of updated coarse flows for performing backward warping on the input frames separately.
- Localization Distillation for Dense Object Detection208
This paper reformulates the knowledge distillation process on localization to present a novel localization distillation method which can efficiently transfer the localization knowledge from the teacher to the student and heuristically introduces the concept of valuable localization region that can aid to selectively distill the semantic and localization knowledge for a certain region.
- CamoFormer: Masked Separable Attention for Camouflaged Object Detection198
Inspired by the multi-head self-attention in Transformers, a simple masked separable attention (MSA) for camouflaged object detection is presented and proposed to capture high-resolution semantic representations progressively based on a simple top-down decoder with the proposed MSA to attain precise segmentation results.
- L2G: A Simple Local-to-Global Knowledge Transfer Framework for Weakly Supervised Semantic Segmentation192
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.
- CrossKD: Cross-Head Knowledge Distillation for Object Detection160
This paper presents a general and effective prediction mimicking distillation scheme, called CrossKD, which delivers the intermediate features of the student's detection head to the teacher's detection head, and the resulting crosshead predictions are forced to mimic the teacher's predictions.
- Dynamic Feature Integration for Simultaneous Detection of Salient Object, Edge, and Skeleton160
Though these tasks are naturally quite different, the proposed network can work well on all of them and even perform better than current single-purpose state-of-the-art methods.
- DFormer: Rethinking RGBD Representation Learning for Semantic Segmentation144
Experimental results show that the DFormer achieves new state-of-the-art performance on these two tasks with less than half of the computational cost of the current best methods on two RGB-D semantic segmentation datasets and five RGB-D salient object detection datasets.
- CorrMatch: Label Propagation via Correlation Matching for Semi-Supervised Semantic Segmentation109
This paper proposes to conduct pixel propagation by modeling the pairwise similarities of pixels to spread the high-confidence pixels and dig out more, and performs region propagation to enhance the pseudo labels with accurate class-agnostic masks extracted from the correlation maps.
- Localization Distillation for Object Detection102
This paper presents a novel localization distillation (LD) method which can efficiently transfer the localization knowledge from the teacher to the student and introduces the concept of valuable localization region that can aid to selectively distill the classification and localization knowledge for a certain region.
- Associating Inter-image Salient Instances for Weakly Supervised Semantic Segmentation101
This paper uses an instance-level salient object detector to automatically generate salient instances (candidate objects) for training images, and proposes a graph-partitioning-based clustering algorithm that outperforms state-of-the-art weakly supervised alternatives by a large margin.
- S4Net: Single stage salient-instance segmentation97
This paper designs a single stage salient instance segmentation framework, with a novel segmentation branch that regards not only local context inside each detection window but also the surrounding context, enabling us to distinguish instances in the same scope even with partial occlusion.
- Multi-Task Dense Prediction via Mixture of Low-Rank Experts96
A novel decoder-focused method for multitask dense prediction, called Mixture-of-Low-Rank-Experts (MLoRE), which adds a generic convolution path to the original MoE structure, where each task feature can go through this path for explicit parameter sharing.
- Intelligent Visual Media Processing: When Graphics Meets Vision87
Recent research on how computer vision techniques benefit computer graphics techniques and vice versa is surveyed, and research on analysis, manipulation, synthesis, and interaction is covered.
- 79
- DFormerv2: Geometry Self-Attention for RGBD Semantic Segmentation77
This paper investigates a new way to learn RGBD feature representations and presents DFormerv2, a strong RGBD encoder that explicitly uses depth maps as geometry priors rather than encoding depth information with neural networks.
- MCANet: Medical Image Segmentation with Multi-scale Cross-axis Attention76
This paper proposes the multi-scale cross-axis attention (MCA) mechanism and develops the network architecture, termed MCANet, which outperforms previous heavyweight approaches across four challenging tasks: skin lesion segmentation, nuclei segmentation, abdominal multi-organ segmentation, and polyp segmentation.
- HFS: Hierarchical Feature Selection for Efficient Image Segmentation69
This paper proposes a real-time system, Hierarchical Feature Selection (HFS), that performs image segmentation at a speed of 50 frames-per-second, and can be used in a variety computer vision tasks that are built on top ofimage segmentation and superpixel extraction.
- FLIC: Fast linear iterative clustering with active search56
This paper proposes a novel search algorithm called “active search” which explicitly considers neighbor continuity, and designs a back-and-forth traversal strategy and a joint assignment and update step to speed up the algorithm.
- 52
- 51
- Adversarial Domain Adaptation With Prototype-Based Normalized Output Conditioner44
This work improves concatenation conditioning by normalizing the output predictions to have the same norm of features, and term the derived method as Normalized OutpUt coNditioner (NOUN), and proposes to condition the cross-domain feature alignment in the prototype space rather than in the output space.
- 40
- SRFormerV2: Taking a Closer Look at Permuted Self-Attention for Image Super-Resolution39
This paper presents SRFormer, a simple but novel method that can enjoy the benefit of large window self-attention but introduces even less computational burden, and attempts to scale up the model by further enlarging the window size and channel numbers to explore the potential of Transformer-based models.
- Referring Camouflaged Object Detection39
This work develops a simple but strong dual-branch framework, dubbed R2CNet, with a reference branch embedding the common representations of target objects from referring images and a segmentation branch identifying and segmenting camouflaged objects under the guidance of the common representations.
- 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.
- Docopilot: Improving Multimodal Models for Document-Level Understanding38
This work presents a high-quality document-level dataset, Doc-750K, designed to support in-depth understanding of multimodal documents, and develops a native multimodal model, Docopilot, which can accurately handle document-level dependencies without relying on RAG.
- 37
- SpinNet: Spinning convolutional network for lane boundary detection35
A novel spinning convolution layer and a brand-new lane parameterization branch are designed and tested to detect lane boundaries from a global perspective, and it is shown that the proposed SpinNet outperforms state-of-the-art methods.
- Cascade-CLIP: Cascaded Vision-Language Embeddings Alignment for Zero-Shot Semantic Segmentation33
This work introduces a series of independent decoders to align the multi-level visual features with the text embeddings in a cascaded way, forming a novel but simple framework named Cascade-CLIP that achieves superior zero-shot performance on segmentation benchmarks.
- Sora Generates Videos with Stunning Geometrical Consistency31
This paper introduces a new benchmark that assesses the quality of the generated videos based on their adherence to real-world physics principles, and employs a method that transforms the generated videos into 3D models, leveraging the premise that the accuracy of 3D reconstruction is heavily contingent on the video quality.
- FMNet: Frequency-Aware Modulation Network for SDR-to-HDR Translation29
A frequency-aware modulation block that can dynamically modulate the features according to its frequency-domain responses is designed to enhance the contrast in a frequency-adaptive way for SDR-to-HDR translation and reduce the structural distortions and artifacts in the translated low-frequency regions.
- 28
- Three Birds One Stone: A Unified Framework for Salient Object Segmentation, Edge Detection and Skeleton Extraction.24
This paper introduces a horizontal cascade, each component of which is densely connected to the outputs of previous component, which allows for effectively exploiting features across different levels hierarchically to effectively address the multiple pixel-wise binary regression tasks.
- 22
- 21
- TAR3D: Creating High-Quality 3D Assets Via Next-Part Prediction20
TAR3D is presented, a novel framework that consists of a 3D-aware Vector Quantized-Variational AutoEncoder and a Generative Pre-trained Transformer to generate high-quality 3D assets to migrate the multimodal unification and promising learning capabilities of the next-token prediction paradigm to conditional 3D object generation.
- TeMO: Towards Text-Driven 3D Stylization for Multi-Object Meshes20
A novel framework, dubbed TeMO, to parse multi-object 3D scenes and edit their styles under the contrast supervision at multiple levels and develops a Cross-Grained Contrast (CGC) supervision system, where a fine-grained loss between the words in the textual description and the randomly rendered images are constructed to complement the coarse-grained loss.
- 19
- A Glimpse to Compress: Dynamic Visual Token Pruning for Large Vision-Language Models18
A dynamic pruning framework inspired by human cognition that takes a data-driven “glimpse” and prunes irrelevant visual tokens in a single forward pass before answer generation and reduces the time and memory costs of fine-tuning with negligible performance loss.
- AR-1-to-3: Single Image to Consistent 3D Object via Next-View Prediction17
AR-1-to-3 is proposed, a novel next-view prediction paradigm based on diffusion models that first generates views close to the input views, which are then utilized as contextual information to progressively synthesize farther views, and a stacked local feature encoding strategy and an LSTMbased global feature encoding strategy.
- 16
- KAC: Kolmogorov-Arnold Classifier for Continual Learning16
This paper introduces the Kolmogorov-Arnold Classifier (KAC), a novel classifier developed for continual learning based on the KAN structure, and explores the impact of KAN’s spline functions and introduces Radial Basis functions for improved compatibility with continual learning.
- A Decoupled Spatio-Temporal Framework for Skeleton-based Action Segmentation16
Joint-decoupled temporal modeling is proposed, which employs independent trainable weights to capture distinctive temporal features of each joint, inspired by the fact that different joints move at different speeds.
- CMAE-V: Contrastive Masked Autoencoders for Video Action Recognition16
This work shows that CMAE also trivially generalizes well on video action recognition without modifying the architecture and the loss criterion, by directly replacing the original pixel shift with the temporal shift.
- Revisiting Efficient Semantic Segmentation: Learning Offsets for Better Spatial and Class Feature Alignment14
This work proposes a coupled dual-branch offset learning paradigm that explicitly learns feature and class offsets to dynamically refine both class representations and spatial image features and constructs an efficient semantic segmentation network, OffSeg.
- 14
- PANDA: Prototypical Unsupervised Domain Adaptation.14
This work attempts to calibrate the noisy pseudo labels with prototypes, and develops an unified framework termed PrototypicAl uNsupervised Domain Adaptation (PANDA) for UDA, which achieves state-of-the-art or competitive results on multiple UDA benchmarks including both object recognition and semantic segmentation tasks.
- 14
- Dual Frequency Transformer for Efficient SDR-to-HDR Translation12
This paper is among the first to investigate the performance of Transformers for SDR-to-HDR translation and proposes a dual frequency attention (DFA), which leverages the self-attention mechanism to separately encode the low-frequency structural information and high-frequency detail information.
- Zone Evaluation: Revealing Spatial Bias in Object Detection12
A new zone evaluation protocol is presented, extending from the traditional evaluation to a more generalized one, which measures the detection performance over zones, yielding a series of Zone Precisions (ZPs), and finds that the key lies in the human-imperceptible divergence in data patterns between objects in different zones, thus eventually forming a visible performance gap between the zones.
- 11
- Unbiased Region-Language Alignment for Open-Vocabulary Dense Prediction11
DenseVLM is proposed, a framework designed to learn unbiased region-language alignment from powerful pretrained VLM representations that can directly replace the original VLM in open-vocabulary object detection and image segmentation methods, leading to notable performance improvements.
- 10
- Towards Universal Video MLLMs with Attribute-Structured and Quality-Verified Instructions9
Experiments across seven benchmarks show that ASID-Captioner improves fine-grained caption quality while reducing hallucinations and improving instruction following and achieves state-of-the-art performance among open-source models and is competitive with Gemini-3-Pro.
- 9
- 9
- 8
- Camouflaged Object Detection with Adaptive Partition and Background Retrieval8
This paper presents a novel learning framework for camouflaged object detection, called AdaptCOD, which decouples the detection process into three parts, namely localization, segmentation, and retrieval.
- 8
- 8
- 8
- 7
- Delving Deeper into Data Scaling in Masked Image Modeling7
The study reveals that MIM can be viewed as an effective method to improve the model capacity when the scale of the training data is relatively small and MIM pre- training is data-agnostic under most scenarios, which means that the strategy of sampling pre-training data is non-critical.
- 7
- 6
- Re-Aligning Language to Visual Objects with an Agentic Workflow6
This work proposes an agentic workflow controlled by an LLM to re-align language to visual objects via adaptively adjusting image and text prompts and reveals a potential to preserve data quality along with scaling up data quantity, which further improves LOD performance from a data-alignment perspective.
- 6
- Traffic Scene Parsing Through the TSP6K Dataset6
A detail refining decoder is proposed for scene parsing, which recovers the details of different semantic regions in traffic scenes owing to the proposed TSP6K dataset, and experiments show its effectiveness in parsing the traffic monitoring scenes.
- 6
- Direct 3D-Aware Object Insertion via Decomposed Visual Proxies5
DI (Decomposed Injection for Reference Composition and Target-integration), a novel framework that integrates interactive pose manipulation with high-fidelity 2D image synthesis to enable pose-controllable object insertion and outperforms previous methods in both geometric controllability and visual quality.
- 5
- 5
- ControlSR: Taming Diffusion Models for Consistent Real-World Image Super Resolution4
ClearSR is presented, a new method that can better take advantage of latent low-resolution image (LR) embeddings for diffusion-based real-world image super-resolution (Real-ISR) and can generate more consistent SR results with LR images than existing methods.
- 4
- 4
- 3
- 3
- Deep negative correlation classification2
This work proposes deep negative correlation classification (DNCC), in which the accuracy and diversity trade-off is systematically controlled by decomposing the loss function seamlessly into individual accuracy and the “correlation” between individual models and the ensemble.
- 2
- 2
- Mixture of Style Experts for Diverse Image Stylization1
StyleExpert is introduced, a semantic-aware framework based on the Mixture of Experts (MoE), which employs a unified style encoder, trained on the large-scale dataset of content-style-stylized triplets, to embed diverse styles into a consistent latent space.
- 1
- 1
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- –
- Towards Spatial Equilibrium Object Detection–
This paper proposes a spatial equilibrium label assignment (SELA) to alleviate the spatial disequilibrium problem by injecting the prior spatial weight into the optimization process of detectors and challenges the conventional sense of object detectors and shows the indispensability of spatial equilibrium.
- –
- –
- –
- –
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 2026-10-13. 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.